librarian-bots/model_cards_with_metadata · Datasets at Hugging Face

modelId string	author string	last_modified timestamp[us, tz=UTC]	downloads int64	likes int64	library_name string	tags sequence	pipeline_tag string	createdAt timestamp[us, tz=UTC]	card string
nhunglaaaaaaa/fccd41c3-68c9-4af6-9694-bcf592003ec3	nhunglaaaaaaa	2025-01-20T23:58:40Z	8	0	peft	[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "base_model:NousResearch/Hermes-2-Pro-Mistral-7B", "base_model:adapter:NousResearch/Hermes-2-Pro-Mistral-7B", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:37:37Z	--- library_name: peft license: apache-2.0 base_model: NousResearch/Hermes-2-Pro-Mistral-7B tags: - axolotl - generated_from_trainer model-index: - name: fccd41c3-68c9-4af6-9694-bcf592003ec3 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/Hermes-2-Pro-Mistral-7B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 670127402f937a76_train_data.json ds_type: json format: custom path: /workspace/input_data/670127402f937a76_train_data.json type: field_input: content field_instruction: aspect field_output: sentiment format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nhunglaaaaaaa/fccd41c3-68c9-4af6-9694-bcf592003ec3 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/670127402f937a76_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: e13dcc00-4d7d-439f-bbb5-ed9061820333 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: e13dcc00-4d7d-439f-bbb5-ed9061820333 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # fccd41c3-68c9-4af6-9694-bcf592003ec3 This model is a fine-tuned version of [NousResearch/Hermes-2-Pro-Mistral-7B](https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2271 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.9569 \| 0.6070 \| 200 \| 0.2271 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
ClarenceDan/eb986324-3dd5-4e2b-868e-a31fb3ad18d2	ClarenceDan	2025-01-20T23:57:26Z	8	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "region:us" ]	null	2025-01-20T23:41:06Z	--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: eb986324-3dd5-4e2b-868e-a31fb3ad18d2 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 854bca96bed40197_train_data.json ds_type: json format: custom path: /workspace/input_data/854bca96bed40197_train_data.json type: field_input: state_before field_instruction: tactic field_output: state_after format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: ClarenceDan/eb986324-3dd5-4e2b-868e-a31fb3ad18d2 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/854bca96bed40197_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: cff9d1c5-a847-4707-b347-d0451baf6b24 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: cff9d1c5-a847-4707-b347-d0451baf6b24 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # eb986324-3dd5-4e2b-868e-a31fb3ad18d2 This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.1671 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 2.1411 \| 0.0000 \| 1 \| 2.8415 \| \| 0.8263 \| 0.0001 \| 3 \| 2.8217 \| \| 0.3436 \| 0.0002 \| 6 \| 2.6084 \| \| 0.4392 \| 0.0003 \| 9 \| 2.1671 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
lesso11/1db4e142-5823-4549-b311-c5325f577241	lesso11	2025-01-20T23:55:46Z	8	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:48:03Z	--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: 1db4e142-5823-4549-b311-c5325f577241 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: true chat_template: llama3 datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso11/1db4e142-5823-4549-b311-c5325f577241 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 1db4e142-5823-4549-b311-c5325f577241 This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.0004 \| 1 \| nan \| \| 0.0 \| 0.0021 \| 5 \| nan \| \| 0.0 \| 0.0042 \| 10 \| nan \| \| 0.0 \| 0.0063 \| 15 \| nan \| \| 0.0 \| 0.0085 \| 20 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
joboffer/c80d23f2-e3e7-4dab-a9fa-432251a8e9d7	joboffer	2025-01-20T23:54:50Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "region:us" ]	null	2025-01-20T23:48:23Z	--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: c80d23f2-e3e7-4dab-a9fa-432251a8e9d7 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: joboffer/c80d23f2-e3e7-4dab-a9fa-432251a8e9d7 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 80GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 2048 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # c80d23f2-e3e7-4dab-a9fa-432251a8e9d7 This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0004 \| 1 \| nan \| \| 0.0 \| 0.0021 \| 5 \| nan \| \| 0.0 \| 0.0042 \| 10 \| nan \| \| 0.0 \| 0.0063 \| 15 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
nadejdatarabukina/c739481e-a786-4a3a-b2b4-3cd78c486811	nadejdatarabukina	2025-01-20T23:54:08Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "region:us" ]	null	2025-01-20T23:48:13Z	--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: c739481e-a786-4a3a-b2b4-3cd78c486811 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: null eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: nadejdatarabukina/c739481e-a786-4a3a-b2b4-3cd78c486811 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 75GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 10 weight_decay: 0.01 xformers_attention: true ``` </details><br> # c739481e-a786-4a3a-b2b4-3cd78c486811 This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0004 \| 1 \| nan \| \| 0.0 \| 0.0021 \| 5 \| nan \| \| 0.0 \| 0.0042 \| 10 \| nan \| \| 0.0 \| 0.0063 \| 15 \| nan \| \| 0.0 \| 0.0085 \| 20 \| nan \| \| 0.0 \| 0.0106 \| 25 \| nan \| \| 0.0 \| 0.0127 \| 30 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
LHRuig/aireduxreal	LHRuig	2025-01-20T23:53:34Z	22	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:53:25Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # aireduxreal <Gallery /> ## Model description aireduxreal lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/aireduxreal/tree/main) them in the Files & versions tab.
lesso09/f177f676-835b-4da0-b0b0-ede96e7966ba	lesso09	2025-01-20T23:52:14Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:princeton-nlp/Sheared-LLaMA-1.3B", "base_model:adapter:princeton-nlp/Sheared-LLaMA-1.3B", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:50:43Z	--- library_name: peft license: apache-2.0 base_model: princeton-nlp/Sheared-LLaMA-1.3B tags: - axolotl - generated_from_trainer model-index: - name: f177f676-835b-4da0-b0b0-ede96e7966ba results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: princeton-nlp/Sheared-LLaMA-1.3B bf16: true chat_template: llama3 datasets: - data_files: - 641edd60ca44ac19_train_data.json ds_type: json format: custom path: /workspace/input_data/641edd60ca44ac19_train_data.json type: field_input: tokens field_instruction: sentence field_output: corrupted format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso09/f177f676-835b-4da0-b0b0-ede96e7966ba hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/641edd60ca44ac19_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ab7df99e-418b-42ee-9175-b39bc5d0f0ce wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ab7df99e-418b-42ee-9175-b39bc5d0f0ce warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # f177f676-835b-4da0-b0b0-ede96e7966ba This model is a fine-tuned version of [princeton-nlp/Sheared-LLaMA-1.3B](https://huggingface.co/princeton-nlp/Sheared-LLaMA-1.3B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.0026 \| 1 \| nan \| \| 0.0 \| 0.0130 \| 5 \| nan \| \| 0.0 \| 0.0260 \| 10 \| nan \| \| 0.0 \| 0.0390 \| 15 \| nan \| \| 0.0 \| 0.0520 \| 20 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
MayBashendy/ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k18_task5_organization	MayBashendy	2025-01-20T23:52:05Z	7	0	transformers	[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-classification	2025-01-20T16:11:17Z	--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k18_task5_organization results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k18_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.0523 - Qwk: 0.4790 - Mse: 1.0523 - Rmse: 1.0258 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Qwk \| Mse \| Rmse \| \|:-------------:\|:------:\|:----:\|:---------------:\|:-------:\|:------:\|:------:\| \| No log \| 0.0345 \| 2 \| 3.8827 \| -0.0151 \| 3.8827 \| 1.9705 \| \| No log \| 0.0690 \| 4 \| 2.3503 \| -0.0189 \| 2.3503 \| 1.5331 \| \| No log \| 0.1034 \| 6 \| 3.3976 \| -0.0343 \| 3.3976 \| 1.8433 \| \| No log \| 0.1379 \| 8 \| 4.3908 \| 0.0171 \| 4.3908 \| 2.0954 \| \| No log \| 0.1724 \| 10 \| 2.9131 \| -0.0295 \| 2.9131 \| 1.7068 \| \| No log \| 0.2069 \| 12 \| 1.3737 \| 0.0894 \| 1.3737 \| 1.1720 \| \| No log \| 0.2414 \| 14 \| 1.0962 \| 0.2416 \| 1.0962 \| 1.0470 \| \| No log \| 0.2759 \| 16 \| 1.0484 \| 0.3449 \| 1.0484 \| 1.0239 \| \| No log \| 0.3103 \| 18 \| 1.1433 \| 0.3927 \| 1.1433 \| 1.0692 \| \| No log \| 0.3448 \| 20 \| 1.5454 \| 0.1429 \| 1.5454 \| 1.2431 \| \| No log \| 0.3793 \| 22 \| 1.9269 \| 0.1296 \| 1.9269 \| 1.3881 \| \| No log \| 0.4138 \| 24 \| 2.1738 \| 0.1465 \| 2.1738 \| 1.4744 \| \| No log \| 0.4483 \| 26 \| 2.2513 \| 0.1221 \| 2.2513 \| 1.5004 \| \| No log \| 0.4828 \| 28 \| 1.6165 \| 0.2351 \| 1.6165 \| 1.2714 \| \| No log \| 0.5172 \| 30 \| 1.1390 \| 0.2560 \| 1.1390 \| 1.0672 \| \| No log \| 0.5517 \| 32 \| 0.9792 \| 0.2526 \| 0.9792 \| 0.9895 \| \| No log \| 0.5862 \| 34 \| 1.0101 \| 0.3066 \| 1.0101 \| 1.0050 \| \| No log \| 0.6207 \| 36 \| 1.1235 \| 0.4012 \| 1.1235 \| 1.0600 \| \| No log \| 0.6552 \| 38 \| 1.1648 \| 0.3542 \| 1.1648 \| 1.0793 \| \| No log \| 0.6897 \| 40 \| 1.1022 \| 0.3243 \| 1.1022 \| 1.0499 \| \| No log \| 0.7241 \| 42 \| 1.1218 \| 0.3243 \| 1.1218 \| 1.0591 \| \| No log \| 0.7586 \| 44 \| 1.0679 \| 0.2977 \| 1.0679 \| 1.0334 \| \| No log \| 0.7931 \| 46 \| 1.0611 \| 0.2567 \| 1.0611 \| 1.0301 \| \| No log \| 0.8276 \| 48 \| 0.9578 \| 0.3691 \| 0.9578 \| 0.9787 \| \| No log \| 0.8621 \| 50 \| 0.8676 \| 0.2770 \| 0.8676 \| 0.9315 \| \| No log \| 0.8966 \| 52 \| 0.8749 \| 0.2770 \| 0.8749 \| 0.9354 \| \| No log \| 0.9310 \| 54 \| 0.8883 \| 0.3498 \| 0.8883 \| 0.9425 \| \| No log \| 0.9655 \| 56 \| 0.9563 \| 0.3250 \| 0.9563 \| 0.9779 \| \| No log \| 1.0 \| 58 \| 1.1407 \| 0.3283 \| 1.1407 \| 1.0681 \| \| No log \| 1.0345 \| 60 \| 1.0584 \| 0.2711 \| 1.0584 \| 1.0288 \| \| No log \| 1.0690 \| 62 \| 0.9629 \| 0.3129 \| 0.9629 \| 0.9813 \| \| No log \| 1.1034 \| 64 \| 1.0630 \| 0.3902 \| 1.0630 \| 1.0310 \| \| No log \| 1.1379 \| 66 \| 1.1496 \| 0.2471 \| 1.1496 \| 1.0722 \| \| No log \| 1.1724 \| 68 \| 1.1000 \| 0.3024 \| 1.1000 \| 1.0488 \| \| No log \| 1.2069 \| 70 \| 0.9757 \| 0.3326 \| 0.9757 \| 0.9878 \| \| No log \| 1.2414 \| 72 \| 0.9358 \| 0.3198 \| 0.9358 \| 0.9674 \| \| No log \| 1.2759 \| 74 \| 0.9934 \| 0.1794 \| 0.9934 \| 0.9967 \| \| No log \| 1.3103 \| 76 \| 1.0084 \| 0.2188 \| 1.0084 \| 1.0042 \| \| No log \| 1.3448 \| 78 \| 1.0067 \| 0.3713 \| 1.0067 \| 1.0033 \| \| No log \| 1.3793 \| 80 \| 0.9656 \| 0.3455 \| 0.9656 \| 0.9827 \| \| No log \| 1.4138 \| 82 \| 0.9832 \| 0.3198 \| 0.9832 \| 0.9915 \| \| No log \| 1.4483 \| 84 \| 1.0364 \| 0.2795 \| 1.0364 \| 1.0180 \| \| No log \| 1.4828 \| 86 \| 1.0609 \| 0.3063 \| 1.0609 \| 1.0300 \| \| No log \| 1.5172 \| 88 \| 1.0794 \| 0.2582 \| 1.0794 \| 1.0389 \| \| No log \| 1.5517 \| 90 \| 1.0631 \| 0.3860 \| 1.0631 \| 1.0311 \| \| No log \| 1.5862 \| 92 \| 1.0703 \| 0.3915 \| 1.0703 \| 1.0346 \| \| No log \| 1.6207 \| 94 \| 1.2684 \| 0.3070 \| 1.2684 \| 1.1262 \| \| No log \| 1.6552 \| 96 \| 1.4565 \| 0.3333 \| 1.4565 \| 1.2068 \| \| No log \| 1.6897 \| 98 \| 1.3804 \| 0.3318 \| 1.3804 \| 1.1749 \| \| No log \| 1.7241 \| 100 \| 1.1446 \| 0.2837 \| 1.1446 \| 1.0698 \| \| No log \| 1.7586 \| 102 \| 1.0397 \| 0.3133 \| 1.0397 \| 1.0197 \| \| No log \| 1.7931 \| 104 \| 1.0443 \| 0.4373 \| 1.0443 \| 1.0219 \| \| No log \| 1.8276 \| 106 \| 1.0965 \| 0.4392 \| 1.0965 \| 1.0471 \| \| No log \| 1.8621 \| 108 \| 1.0463 \| 0.4206 \| 1.0463 \| 1.0229 \| \| No log \| 1.8966 \| 110 \| 1.0460 \| 0.3482 \| 1.0460 \| 1.0228 \| \| No log \| 1.9310 \| 112 \| 1.0853 \| 0.3711 \| 1.0853 \| 1.0418 \| \| No log \| 1.9655 \| 114 \| 1.1043 \| 0.3758 \| 1.1043 \| 1.0509 \| \| No log \| 2.0 \| 116 \| 1.1332 \| 0.3718 \| 1.1332 \| 1.0645 \| \| No log \| 2.0345 \| 118 \| 1.1377 \| 0.3917 \| 1.1377 \| 1.0666 \| \| No log \| 2.0690 \| 120 \| 1.1435 \| 0.3648 \| 1.1435 \| 1.0693 \| \| No log \| 2.1034 \| 122 \| 1.1013 \| 0.4232 \| 1.1013 \| 1.0494 \| \| No log \| 2.1379 \| 124 \| 0.9759 \| 0.3607 \| 0.9759 \| 0.9879 \| \| No log \| 2.1724 \| 126 \| 0.8873 \| 0.3820 \| 0.8873 \| 0.9420 \| \| No log \| 2.2069 \| 128 \| 0.8663 \| 0.3625 \| 0.8663 \| 0.9307 \| \| No log \| 2.2414 \| 130 \| 0.8625 \| 0.4089 \| 0.8625 \| 0.9287 \| \| No log \| 2.2759 \| 132 \| 0.8833 \| 0.4056 \| 0.8833 \| 0.9398 \| \| No log \| 2.3103 \| 134 \| 0.9124 \| 0.4313 \| 0.9124 \| 0.9552 \| \| No log \| 2.3448 \| 136 \| 0.9654 \| 0.3286 \| 0.9654 \| 0.9826 \| \| No log \| 2.3793 \| 138 \| 1.1335 \| 0.3687 \| 1.1335 \| 1.0647 \| \| No log \| 2.4138 \| 140 \| 1.0551 \| 0.3918 \| 1.0551 \| 1.0272 \| \| No log \| 2.4483 \| 142 \| 1.0378 \| 0.4737 \| 1.0378 \| 1.0187 \| \| No log \| 2.4828 \| 144 \| 1.1433 \| 0.4264 \| 1.1433 \| 1.0693 \| \| No log \| 2.5172 \| 146 \| 1.0562 \| 0.4503 \| 1.0562 \| 1.0277 \| \| No log \| 2.5517 \| 148 \| 1.0218 \| 0.2694 \| 1.0218 \| 1.0109 \| \| No log \| 2.5862 \| 150 \| 0.9906 \| 0.3514 \| 0.9906 \| 0.9953 \| \| No log \| 2.6207 \| 152 \| 0.9382 \| 0.3224 \| 0.9382 \| 0.9686 \| \| No log \| 2.6552 \| 154 \| 0.9202 \| 0.2941 \| 0.9202 \| 0.9592 \| \| No log \| 2.6897 \| 156 \| 0.9203 \| 0.3357 \| 0.9203 \| 0.9593 \| \| No log \| 2.7241 \| 158 \| 0.9347 \| 0.4272 \| 0.9347 \| 0.9668 \| \| No log \| 2.7586 \| 160 \| 0.9881 \| 0.4273 \| 0.9881 \| 0.9940 \| \| No log \| 2.7931 \| 162 \| 1.0588 \| 0.4309 \| 1.0588 \| 1.0290 \| \| No log \| 2.8276 \| 164 \| 1.1410 \| 0.4973 \| 1.1410 \| 1.0682 \| \| No log \| 2.8621 \| 166 \| 1.3562 \| 0.2730 \| 1.3562 \| 1.1646 \| \| No log \| 2.8966 \| 168 \| 1.4988 \| 0.2789 \| 1.4988 \| 1.2242 \| \| No log \| 2.9310 \| 170 \| 1.3422 \| 0.2230 \| 1.3422 \| 1.1585 \| \| No log \| 2.9655 \| 172 \| 1.1816 \| 0.4020 \| 1.1816 \| 1.0870 \| \| No log \| 3.0 \| 174 \| 1.1237 \| 0.3924 \| 1.1237 \| 1.0601 \| \| No log \| 3.0345 \| 176 \| 1.1163 \| 0.3843 \| 1.1163 \| 1.0566 \| \| No log \| 3.0690 \| 178 \| 1.2763 \| 0.4032 \| 1.2763 \| 1.1297 \| \| No log \| 3.1034 \| 180 \| 1.3986 \| 0.3243 \| 1.3986 \| 1.1826 \| \| No log \| 3.1379 \| 182 \| 1.2453 \| 0.3345 \| 1.2453 \| 1.1159 \| \| No log \| 3.1724 \| 184 \| 0.9703 \| 0.4161 \| 0.9703 \| 0.9850 \| \| No log \| 3.2069 \| 186 \| 0.9294 \| 0.3804 \| 0.9294 \| 0.9641 \| \| No log \| 3.2414 \| 188 \| 0.9516 \| 0.4845 \| 0.9516 \| 0.9755 \| \| No log \| 3.2759 \| 190 \| 0.9303 \| 0.4223 \| 0.9303 \| 0.9645 \| \| No log \| 3.3103 \| 192 \| 1.0026 \| 0.4196 \| 1.0026 \| 1.0013 \| \| No log \| 3.3448 \| 194 \| 1.1127 \| 0.4976 \| 1.1127 \| 1.0548 \| \| No log \| 3.3793 \| 196 \| 1.0237 \| 0.4568 \| 1.0237 \| 1.0118 \| \| No log \| 3.4138 \| 198 \| 0.9068 \| 0.3454 \| 0.9068 \| 0.9523 \| \| No log \| 3.4483 \| 200 \| 0.8817 \| 0.4118 \| 0.8817 \| 0.9390 \| \| No log \| 3.4828 \| 202 \| 0.9082 \| 0.4729 \| 0.9082 \| 0.9530 \| \| No log \| 3.5172 \| 204 \| 0.9099 \| 0.4581 \| 0.9099 \| 0.9539 \| \| No log \| 3.5517 \| 206 \| 0.8902 \| 0.4350 \| 0.8902 \| 0.9435 \| \| No log \| 3.5862 \| 208 \| 0.9101 \| 0.4215 \| 0.9101 \| 0.9540 \| \| No log \| 3.6207 \| 210 \| 0.9443 \| 0.4807 \| 0.9443 \| 0.9717 \| \| No log \| 3.6552 \| 212 \| 0.9715 \| 0.4369 \| 0.9715 \| 0.9856 \| \| No log \| 3.6897 \| 214 \| 0.9362 \| 0.3792 \| 0.9362 \| 0.9676 \| \| No log \| 3.7241 \| 216 \| 0.9121 \| 0.3742 \| 0.9121 \| 0.9550 \| \| No log \| 3.7586 \| 218 \| 0.9766 \| 0.4585 \| 0.9766 \| 0.9882 \| \| No log \| 3.7931 \| 220 \| 1.1081 \| 0.4787 \| 1.1081 \| 1.0527 \| \| No log \| 3.8276 \| 222 \| 1.1726 \| 0.4301 \| 1.1726 \| 1.0829 \| \| No log \| 3.8621 \| 224 \| 1.0732 \| 0.4585 \| 1.0732 \| 1.0360 \| \| No log \| 3.8966 \| 226 \| 0.9207 \| 0.4230 \| 0.9207 \| 0.9595 \| \| No log \| 3.9310 \| 228 \| 0.8838 \| 0.4114 \| 0.8838 \| 0.9401 \| \| No log \| 3.9655 \| 230 \| 0.9041 \| 0.4230 \| 0.9041 \| 0.9508 \| \| No log \| 4.0 \| 232 \| 0.9400 \| 0.4745 \| 0.9400 \| 0.9696 \| \| No log \| 4.0345 \| 234 \| 0.9597 \| 0.4930 \| 0.9597 \| 0.9796 \| \| No log \| 4.0690 \| 236 \| 0.9359 \| 0.4306 \| 0.9359 \| 0.9674 \| \| No log \| 4.1034 \| 238 \| 0.8728 \| 0.4069 \| 0.8728 \| 0.9342 \| \| No log \| 4.1379 \| 240 \| 0.8607 \| 0.3981 \| 0.8607 \| 0.9277 \| \| No log \| 4.1724 \| 242 \| 0.8876 \| 0.4405 \| 0.8876 \| 0.9421 \| \| No log \| 4.2069 \| 244 \| 0.8922 \| 0.4160 \| 0.8922 \| 0.9445 \| \| No log \| 4.2414 \| 246 \| 0.9119 \| 0.5073 \| 0.9119 \| 0.9549 \| \| No log \| 4.2759 \| 248 \| 0.9036 \| 0.4364 \| 0.9036 \| 0.9506 \| \| No log \| 4.3103 \| 250 \| 0.9250 \| 0.4760 \| 0.9250 \| 0.9618 \| \| No log \| 4.3448 \| 252 \| 0.9678 \| 0.4076 \| 0.9678 \| 0.9838 \| \| No log \| 4.3793 \| 254 \| 0.9431 \| 0.4063 \| 0.9431 \| 0.9711 \| \| No log \| 4.4138 \| 256 \| 0.9360 \| 0.3705 \| 0.9360 \| 0.9675 \| \| No log \| 4.4483 \| 258 \| 0.9044 \| 0.4186 \| 0.9044 \| 0.9510 \| \| No log \| 4.4828 \| 260 \| 0.9198 \| 0.3705 \| 0.9198 \| 0.9590 \| \| No log \| 4.5172 \| 262 \| 1.0117 \| 0.4058 \| 1.0117 \| 1.0058 \| \| No log \| 4.5517 \| 264 \| 1.0365 \| 0.4286 \| 1.0365 \| 1.0181 \| \| No log \| 4.5862 \| 266 \| 0.9587 \| 0.4033 \| 0.9587 \| 0.9791 \| \| No log \| 4.6207 \| 268 \| 0.9625 \| 0.3842 \| 0.9625 \| 0.9811 \| \| No log \| 4.6552 \| 270 \| 1.0778 \| 0.4405 \| 1.0778 \| 1.0382 \| \| No log \| 4.6897 \| 272 \| 1.2736 \| 0.3972 \| 1.2736 \| 1.1286 \| \| No log \| 4.7241 \| 274 \| 1.2229 \| 0.4874 \| 1.2229 \| 1.1059 \| \| No log \| 4.7586 \| 276 \| 1.0118 \| 0.4510 \| 1.0118 \| 1.0059 \| \| No log \| 4.7931 \| 278 \| 0.8990 \| 0.3640 \| 0.8990 \| 0.9482 \| \| No log \| 4.8276 \| 280 \| 0.8903 \| 0.3640 \| 0.8903 \| 0.9435 \| \| No log \| 4.8621 \| 282 \| 0.9633 \| 0.4171 \| 0.9633 \| 0.9815 \| \| No log \| 4.8966 \| 284 \| 1.1899 \| 0.4491 \| 1.1899 \| 1.0908 \| \| No log \| 4.9310 \| 286 \| 1.1471 \| 0.4487 \| 1.1471 \| 1.0710 \| \| No log \| 4.9655 \| 288 \| 0.9659 \| 0.3512 \| 0.9659 \| 0.9828 \| \| No log \| 5.0 \| 290 \| 0.9598 \| 0.3607 \| 0.9598 \| 0.9797 \| \| No log \| 5.0345 \| 292 \| 1.0185 \| 0.3629 \| 1.0185 \| 1.0092 \| \| No log \| 5.0690 \| 294 \| 1.0927 \| 0.3766 \| 1.0927 \| 1.0453 \| \| No log \| 5.1034 \| 296 \| 1.2132 \| 0.4186 \| 1.2132 \| 1.1015 \| \| No log \| 5.1379 \| 298 \| 1.1753 \| 0.3972 \| 1.1753 \| 1.0841 \| \| No log \| 5.1724 \| 300 \| 1.0267 \| 0.3436 \| 1.0267 \| 1.0132 \| \| No log \| 5.2069 \| 302 \| 0.9602 \| 0.3144 \| 0.9602 \| 0.9799 \| \| No log \| 5.2414 \| 304 \| 0.9394 \| 0.3308 \| 0.9394 \| 0.9692 \| \| No log \| 5.2759 \| 306 \| 0.9721 \| 0.3463 \| 0.9721 \| 0.9859 \| \| No log \| 5.3103 \| 308 \| 1.0627 \| 0.3972 \| 1.0627 \| 1.0309 \| \| No log \| 5.3448 \| 310 \| 1.0346 \| 0.4589 \| 1.0346 \| 1.0172 \| \| No log \| 5.3793 \| 312 \| 0.8910 \| 0.3809 \| 0.8910 \| 0.9439 \| \| No log \| 5.4138 \| 314 \| 0.8535 \| 0.3437 \| 0.8535 \| 0.9238 \| \| No log \| 5.4483 \| 316 \| 0.8935 \| 0.3447 \| 0.8935 \| 0.9453 \| \| No log \| 5.4828 \| 318 \| 0.9043 \| 0.3842 \| 0.9043 \| 0.9510 \| \| No log \| 5.5172 \| 320 \| 1.0219 \| 0.4613 \| 1.0219 \| 1.0109 \| \| No log \| 5.5517 \| 322 \| 1.1213 \| 0.4693 \| 1.1213 \| 1.0589 \| \| No log \| 5.5862 \| 324 \| 1.0448 \| 0.4615 \| 1.0448 \| 1.0222 \| \| No log \| 5.6207 \| 326 \| 0.9554 \| 0.4376 \| 0.9554 \| 0.9775 \| \| No log \| 5.6552 \| 328 \| 0.9215 \| 0.4273 \| 0.9215 \| 0.9600 \| \| No log \| 5.6897 \| 330 \| 0.9059 \| 0.3678 \| 0.9059 \| 0.9518 \| \| No log \| 5.7241 \| 332 \| 0.9500 \| 0.3728 \| 0.9500 \| 0.9747 \| \| No log \| 5.7586 \| 334 \| 0.9382 \| 0.3688 \| 0.9382 \| 0.9686 \| \| No log \| 5.7931 \| 336 \| 0.8925 \| 0.2939 \| 0.8925 \| 0.9447 \| \| No log \| 5.8276 \| 338 \| 0.8878 \| 0.4229 \| 0.8878 \| 0.9422 \| \| No log \| 5.8621 \| 340 \| 0.8877 \| 0.4013 \| 0.8877 \| 0.9422 \| \| No log \| 5.8966 \| 342 \| 0.9140 \| 0.3923 \| 0.9140 \| 0.9560 \| \| No log \| 5.9310 \| 344 \| 0.9673 \| 0.4822 \| 0.9673 \| 0.9835 \| \| No log \| 5.9655 \| 346 \| 1.0367 \| 0.4589 \| 1.0367 \| 1.0182 \| \| No log \| 6.0 \| 348 \| 0.9746 \| 0.4792 \| 0.9746 \| 0.9872 \| \| No log \| 6.0345 \| 350 \| 0.8739 \| 0.3535 \| 0.8739 \| 0.9348 \| \| No log \| 6.0690 \| 352 \| 0.8633 \| 0.3572 \| 0.8633 \| 0.9291 \| \| No log \| 6.1034 \| 354 \| 0.8985 \| 0.4677 \| 0.8985 \| 0.9479 \| \| No log \| 6.1379 \| 356 \| 0.9457 \| 0.4449 \| 0.9457 \| 0.9725 \| \| No log \| 6.1724 \| 358 \| 1.0018 \| 0.3975 \| 1.0018 \| 1.0009 \| \| No log \| 6.2069 \| 360 \| 1.0198 \| 0.3728 \| 1.0198 \| 1.0099 \| \| No log \| 6.2414 \| 362 \| 0.9638 \| 0.3624 \| 0.9638 \| 0.9817 \| \| No log \| 6.2759 \| 364 \| 0.9300 \| 0.3772 \| 0.9300 \| 0.9644 \| \| No log \| 6.3103 \| 366 \| 0.9246 \| 0.3860 \| 0.9246 \| 0.9616 \| \| No log \| 6.3448 \| 368 \| 0.9481 \| 0.3877 \| 0.9481 \| 0.9737 \| \| No log \| 6.3793 \| 370 \| 1.0751 \| 0.4497 \| 1.0751 \| 1.0369 \| \| No log \| 6.4138 \| 372 \| 1.0104 \| 0.4400 \| 1.0104 \| 1.0052 \| \| No log \| 6.4483 \| 374 \| 0.8589 \| 0.4003 \| 0.8589 \| 0.9267 \| \| No log \| 6.4828 \| 376 \| 0.8292 \| 0.5179 \| 0.8292 \| 0.9106 \| \| No log \| 6.5172 \| 378 \| 0.8432 \| 0.5071 \| 0.8432 \| 0.9183 \| \| No log \| 6.5517 \| 380 \| 0.8499 \| 0.3601 \| 0.8499 \| 0.9219 \| \| No log \| 6.5862 \| 382 \| 1.0001 \| 0.4694 \| 1.0001 \| 1.0001 \| \| No log \| 6.6207 \| 384 \| 1.1535 \| 0.4681 \| 1.1535 \| 1.0740 \| \| No log \| 6.6552 \| 386 \| 1.1194 \| 0.4592 \| 1.1194 \| 1.0580 \| \| No log \| 6.6897 \| 388 \| 0.9916 \| 0.4228 \| 0.9916 \| 0.9958 \| \| No log \| 6.7241 \| 390 \| 0.9721 \| 0.3065 \| 0.9721 \| 0.9859 \| \| No log \| 6.7586 \| 392 \| 0.9257 \| 0.3159 \| 0.9257 \| 0.9621 \| \| No log \| 6.7931 \| 394 \| 0.8857 \| 0.2416 \| 0.8857 \| 0.9411 \| \| No log \| 6.8276 \| 396 \| 0.8897 \| 0.2951 \| 0.8897 \| 0.9432 \| \| No log \| 6.8621 \| 398 \| 0.9185 \| 0.3029 \| 0.9185 \| 0.9584 \| \| No log \| 6.8966 \| 400 \| 0.9333 \| 0.3134 \| 0.9333 \| 0.9661 \| \| No log \| 6.9310 \| 402 \| 0.9419 \| 0.3207 \| 0.9419 \| 0.9705 \| \| No log \| 6.9655 \| 404 \| 1.0028 \| 0.3917 \| 1.0028 \| 1.0014 \| \| No log \| 7.0 \| 406 \| 1.0369 \| 0.3946 \| 1.0369 \| 1.0183 \| \| No log \| 7.0345 \| 408 \| 1.0417 \| 0.4116 \| 1.0417 \| 1.0206 \| \| No log \| 7.0690 \| 410 \| 1.0446 \| 0.3348 \| 1.0446 \| 1.0221 \| \| No log \| 7.1034 \| 412 \| 1.0095 \| 0.2742 \| 1.0095 \| 1.0047 \| \| No log \| 7.1379 \| 414 \| 1.0162 \| 0.3430 \| 1.0162 \| 1.0081 \| \| No log \| 7.1724 \| 416 \| 0.9666 \| 0.3404 \| 0.9666 \| 0.9832 \| \| No log \| 7.2069 \| 418 \| 0.9279 \| 0.2786 \| 0.9279 \| 0.9633 \| \| No log \| 7.2414 \| 420 \| 0.9377 \| 0.3457 \| 0.9377 \| 0.9683 \| \| No log \| 7.2759 \| 422 \| 0.9122 \| 0.3457 \| 0.9122 \| 0.9551 \| \| No log \| 7.3103 \| 424 \| 0.8928 \| 0.2899 \| 0.8928 \| 0.9449 \| \| No log \| 7.3448 \| 426 \| 0.9117 \| 0.3601 \| 0.9117 \| 0.9549 \| \| No log \| 7.3793 \| 428 \| 0.8795 \| 0.3289 \| 0.8795 \| 0.9378 \| \| No log \| 7.4138 \| 430 \| 0.8721 \| 0.3357 \| 0.8721 \| 0.9339 \| \| No log \| 7.4483 \| 432 \| 0.8763 \| 0.3224 \| 0.8763 \| 0.9361 \| \| No log \| 7.4828 \| 434 \| 0.9084 \| 0.5222 \| 0.9084 \| 0.9531 \| \| No log \| 7.5172 \| 436 \| 0.8770 \| 0.4268 \| 0.8770 \| 0.9365 \| \| No log \| 7.5517 \| 438 \| 0.8508 \| 0.3980 \| 0.8508 \| 0.9224 \| \| No log \| 7.5862 \| 440 \| 0.9223 \| 0.4205 \| 0.9223 \| 0.9604 \| \| No log \| 7.6207 \| 442 \| 1.0067 \| 0.4400 \| 1.0067 \| 1.0034 \| \| No log \| 7.6552 \| 444 \| 0.9343 \| 0.4822 \| 0.9343 \| 0.9666 \| \| No log \| 7.6897 \| 446 \| 0.8419 \| 0.4247 \| 0.8419 \| 0.9175 \| \| No log \| 7.7241 \| 448 \| 0.8061 \| 0.4463 \| 0.8061 \| 0.8979 \| \| No log \| 7.7586 \| 450 \| 0.8104 \| 0.4221 \| 0.8104 \| 0.9002 \| \| No log \| 7.7931 \| 452 \| 0.8290 \| 0.4327 \| 0.8290 \| 0.9105 \| \| No log \| 7.8276 \| 454 \| 0.8036 \| 0.4088 \| 0.8036 \| 0.8964 \| \| No log \| 7.8621 \| 456 \| 0.7561 \| 0.5431 \| 0.7561 \| 0.8695 \| \| No log \| 7.8966 \| 458 \| 0.7412 \| 0.5057 \| 0.7412 \| 0.8609 \| \| No log \| 7.9310 \| 460 \| 0.7442 \| 0.4706 \| 0.7442 \| 0.8627 \| \| No log \| 7.9655 \| 462 \| 0.7870 \| 0.4421 \| 0.7870 \| 0.8871 \| \| No log \| 8.0 \| 464 \| 0.8435 \| 0.4103 \| 0.8435 \| 0.9184 \| \| No log \| 8.0345 \| 466 \| 0.8581 \| 0.4306 \| 0.8581 \| 0.9263 \| \| No log \| 8.0690 \| 468 \| 0.8054 \| 0.4425 \| 0.8054 \| 0.8974 \| \| No log \| 8.1034 \| 470 \| 0.7978 \| 0.4444 \| 0.7978 \| 0.8932 \| \| No log \| 8.1379 \| 472 \| 0.7923 \| 0.3985 \| 0.7923 \| 0.8901 \| \| No log \| 8.1724 \| 474 \| 0.7926 \| 0.4235 \| 0.7926 \| 0.8903 \| \| No log \| 8.2069 \| 476 \| 0.7973 \| 0.4297 \| 0.7973 \| 0.8929 \| \| No log \| 8.2414 \| 478 \| 0.7842 \| 0.3967 \| 0.7842 \| 0.8856 \| \| No log \| 8.2759 \| 480 \| 0.7854 \| 0.3970 \| 0.7854 \| 0.8862 \| \| No log \| 8.3103 \| 482 \| 0.8013 \| 0.4622 \| 0.8013 \| 0.8952 \| \| No log \| 8.3448 \| 484 \| 0.8078 \| 0.4622 \| 0.8078 \| 0.8988 \| \| No log \| 8.3793 \| 486 \| 0.7642 \| 0.5328 \| 0.7642 \| 0.8742 \| \| No log \| 8.4138 \| 488 \| 0.8224 \| 0.4719 \| 0.8224 \| 0.9069 \| \| No log \| 8.4483 \| 490 \| 1.0762 \| 0.5161 \| 1.0762 \| 1.0374 \| \| No log \| 8.4828 \| 492 \| 1.1014 \| 0.5145 \| 1.1014 \| 1.0495 \| \| No log \| 8.5172 \| 494 \| 0.9148 \| 0.4810 \| 0.9148 \| 0.9565 \| \| No log \| 8.5517 \| 496 \| 0.7722 \| 0.4577 \| 0.7722 \| 0.8787 \| \| No log \| 8.5862 \| 498 \| 0.8145 \| 0.5435 \| 0.8145 \| 0.9025 \| \| 0.312 \| 8.6207 \| 500 \| 0.8249 \| 0.4903 \| 0.8249 \| 0.9083 \| \| 0.312 \| 8.6552 \| 502 \| 0.8091 \| 0.4072 \| 0.8091 \| 0.8995 \| \| 0.312 \| 8.6897 \| 504 \| 0.8141 \| 0.3455 \| 0.8141 \| 0.9023 \| \| 0.312 \| 8.7241 \| 506 \| 0.8404 \| 0.3563 \| 0.8404 \| 0.9167 \| \| 0.312 \| 8.7586 \| 508 \| 0.9525 \| 0.4156 \| 0.9525 \| 0.9759 \| \| 0.312 \| 8.7931 \| 510 \| 1.0523 \| 0.4790 \| 1.0523 \| 1.0258 \| ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1
dimasik2987/c4fd30f8-9bed-4199-bfca-cc8c495b627b	dimasik2987	2025-01-20T23:52:01Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "region:us" ]	null	2025-01-20T23:48:29Z	--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: c4fd30f8-9bed-4199-bfca-cc8c495b627b results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: dimasik2987/c4fd30f8-9bed-4199-bfca-cc8c495b627b hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # c4fd30f8-9bed-4199-bfca-cc8c495b627b This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0008 \| 1 \| nan \| \| 0.0 \| 0.0042 \| 5 \| nan \| \| 0.0 \| 0.0085 \| 10 \| nan \| \| 0.0 \| 0.0127 \| 15 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
VERSIL91/8e83e5ba-b92a-4907-bf39-218caf42228a	VERSIL91	2025-01-20T23:51:49Z	8	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:HuggingFaceM4/tiny-random-LlamaForCausalLM", "base_model:adapter:HuggingFaceM4/tiny-random-LlamaForCausalLM", "region:us" ]	null	2025-01-20T23:51:46Z	--- library_name: peft base_model: HuggingFaceM4/tiny-random-LlamaForCausalLM tags: - axolotl - generated_from_trainer model-index: - name: bef97220-cdcf-4144-9f98-4582cf4a902b results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml accelerate_config: dynamo_backend: inductor mixed_precision: bf16 num_machines: 1 num_processes: auto use_cpu: false adapter: lora base_model: HuggingFaceM4/tiny-random-LlamaForCausalLM bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 7da10487b55868a6_train_data.json ds_type: json format: custom path: /workspace/input_data/7da10487b55868a6_train_data.json type: field_instruction: hyps field_output: ref format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 16 gradient_checkpointing: true group_by_length: false hub_model_id: null hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lora_target_modules: - q_proj - v_proj lr_scheduler: cosine max_memory: 0: 70GiB max_steps: 50 micro_batch_size: 2 mlflow_experiment_name: /tmp/7da10487b55868a6_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true quantization_config: llm_int8_enable_fp32_cpu_offload: true load_in_8bit: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer torch_compile: true train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: bef97220-cdcf-4144-9f98-4582cf4a902b wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: bef97220-cdcf-4144-9f98-4582cf4a902b warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # bef97220-cdcf-4144-9f98-4582cf4a902b This model is a fine-tuned version of [HuggingFaceM4/tiny-random-LlamaForCausalLM](https://huggingface.co/HuggingFaceM4/tiny-random-LlamaForCausalLM) on the None dataset. It achieves the following results on the evaluation set: - Loss: 10.3648 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 16 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 50 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 10.3772 \| 0.0002 \| 1 \| 10.3803 \| \| 10.3751 \| 0.0031 \| 13 \| 10.3758 \| \| 10.3722 \| 0.0062 \| 26 \| 10.3686 \| \| 10.3688 \| 0.0093 \| 39 \| 10.3648 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
kostiantynk1205/41925306-0ef0-4197-b10b-98e1dc6ca5d3	kostiantynk1205	2025-01-20T23:51:38Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2-1.5B-Instruct", "base_model:adapter:unsloth/Qwen2-1.5B-Instruct", "license:apache-2.0", "region:us" ]	null	2025-01-20T23:51:09Z	--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2-1.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 41925306-0ef0-4197-b10b-98e1dc6ca5d3 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2-1.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - dd06633aceb12410_train_data.json ds_type: json format: custom path: /workspace/input_data/dd06633aceb12410_train_data.json type: field_instruction: tests field_output: prompt format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: kostiantynk1205/41925306-0ef0-4197-b10b-98e1dc6ca5d3 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/dd06633aceb12410_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 wandb_project: Birthday-SN56-23-Gradients-On-Demand wandb_run: your_name wandb_runid: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 41925306-0ef0-4197-b10b-98e1dc6ca5d3 This model is a fine-tuned version of [unsloth/Qwen2-1.5B-Instruct](https://huggingface.co/unsloth/Qwen2-1.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 6 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.1905 \| 1 \| nan \| \| 0.0 \| 0.3810 \| 2 \| nan \| \| 0.0 \| 0.7619 \| 4 \| nan \| \| 0.0 \| 1.1429 \| 6 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
dimasik87/150164aa-8220-47b5-a42e-e3e1914c42b8	dimasik87	2025-01-20T23:50:30Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "region:us" ]	null	2025-01-20T23:48:18Z	--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: 150164aa-8220-47b5-a42e-e3e1914c42b8 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: dimasik87/150164aa-8220-47b5-a42e-e3e1914c42b8 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # 150164aa-8220-47b5-a42e-e3e1914c42b8 This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0008 \| 1 \| nan \| \| 0.0 \| 0.0042 \| 5 \| nan \| \| 0.0 \| 0.0085 \| 10 \| nan \| \| 0.0 \| 0.0127 \| 15 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
ardaspear/32cc01b9-8751-4d53-b346-be211e673aa6	ardaspear	2025-01-20T23:49:47Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:codellama/CodeLlama-7b-hf", "base_model:adapter:codellama/CodeLlama-7b-hf", "license:llama2", "region:us" ]	null	2025-01-20T22:33:09Z	--- library_name: peft license: llama2 base_model: codellama/CodeLlama-7b-hf tags: - axolotl - generated_from_trainer model-index: - name: 32cc01b9-8751-4d53-b346-be211e673aa6 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: codellama/CodeLlama-7b-hf bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 25670fa5d5514c5b_train_data.json ds_type: json format: custom path: /workspace/input_data/25670fa5d5514c5b_train_data.json type: field_input: facts field_instruction: prompt_serial field_output: hypothesis format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: ardaspear/32cc01b9-8751-4d53-b346-be211e673aa6 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 5.0e-05 load_in_4bit: false load_in_8bit: false local_rank: 0 logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 100 micro_batch_size: 8 mlflow_experiment_name: /tmp/25670fa5d5514c5b_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: techspear-hub wandb_mode: online wandb_name: ac3b7fa4-8d2e-4dc9-b139-aaeea5f132ac wandb_project: Gradients-On-Five wandb_run: your_name wandb_runid: ac3b7fa4-8d2e-4dc9-b139-aaeea5f132ac warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # 32cc01b9-8751-4d53-b346-be211e673aa6 This model is a fine-tuned version of [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.0776 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 100 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0008 \| 1 \| 0.7974 \| \| 0.7688 \| 0.0076 \| 9 \| 0.7061 \| \| 0.3383 \| 0.0152 \| 18 \| 0.3166 \| \| 0.174 \| 0.0228 \| 27 \| 0.1894 \| \| 0.147 \| 0.0305 \| 36 \| 0.1512 \| \| 0.1256 \| 0.0381 \| 45 \| 0.1198 \| \| 0.1727 \| 0.0457 \| 54 \| 0.0931 \| \| 0.0712 \| 0.0533 \| 63 \| 0.0830 \| \| 0.0857 \| 0.0609 \| 72 \| 0.0793 \| \| 0.0679 \| 0.0685 \| 81 \| 0.0780 \| \| 0.039 \| 0.0762 \| 90 \| 0.0777 \| \| 0.0732 \| 0.0838 \| 99 \| 0.0776 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
tuanna08go/db6fb768-4f34-45e9-a10c-21c12445199e	tuanna08go	2025-01-20T23:49:36Z	8	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:codellama/CodeLlama-7b-hf", "base_model:adapter:codellama/CodeLlama-7b-hf", "license:llama2", "region:us" ]	null	2025-01-20T23:09:11Z	--- library_name: peft license: llama2 base_model: codellama/CodeLlama-7b-hf tags: - axolotl - generated_from_trainer model-index: - name: db6fb768-4f34-45e9-a10c-21c12445199e results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: codellama/CodeLlama-7b-hf bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 25670fa5d5514c5b_train_data.json ds_type: json format: custom path: /workspace/input_data/25670fa5d5514c5b_train_data.json type: field_input: facts field_instruction: prompt_serial field_output: hypothesis format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 5 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: tuanna08go/db6fb768-4f34-45e9-a10c-21c12445199e hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 5 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 50 micro_batch_size: 2 mlflow_experiment_name: /tmp/25670fa5d5514c5b_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ac3b7fa4-8d2e-4dc9-b139-aaeea5f132ac wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ac3b7fa4-8d2e-4dc9-b139-aaeea5f132ac warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # db6fb768-4f34-45e9-a10c-21c12445199e This model is a fine-tuned version of [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.5158 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 50 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0002 \| 1 \| 1.2075 \| \| 1.0753 \| 0.0021 \| 10 \| 1.0732 \| \| 0.7292 \| 0.0042 \| 20 \| 0.6758 \| \| 0.5962 \| 0.0063 \| 30 \| 0.5538 \| \| 0.4682 \| 0.0085 \| 40 \| 0.5197 \| \| 0.4209 \| 0.0106 \| 50 \| 0.5158 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
charlesniswander/fruit_freshness_demo	charlesniswander	2025-01-20T23:49:26Z	28	0	null	[ "tensorboard", "safetensors", "vit", "image-classification", "pytorch", "huggingpics", "model-index", "region:us" ]	image-classification	2025-01-20T23:49:14Z	--- tags: - image-classification - pytorch - huggingpics metrics: - accuracy model-index: - name: fruit_freshness_demo results: - task: name: Image Classification type: image-classification metrics: - name: Accuracy type: accuracy value: 0.3483146131038666 --- # fruit_freshness_demo Autogenerated by HuggingPics🤗🖼️ Create your own image classifier for anything by running [the demo on Google Colab](https://colab.research.google.com/github/nateraw/huggingpics/blob/main/HuggingPics.ipynb). Report any issues with the demo at the [github repo](https://github.com/nateraw/huggingpics). ## Example Images #### fresh apple ![fresh apple](images/fresh_apple.jpg) #### fresh banana ![fresh banana](images/fresh_banana.jpg) #### rotten apple ![rotten apple](images/rotten_apple.jpg) #### rotten banana ![rotten banana](images/rotten_banana.jpg)
taopanda-2/a31b520a-ea45-423d-a41b-09df0cfc856c	taopanda-2	2025-01-20T23:49:24Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:06:04Z	--- license: apache-2.0 library_name: peft tags: - axolotl - generated_from_trainer base_model: Qwen/Qwen2.5-0.5B-Instruct model-index: - name: a31b520a-ea45-423d-a41b-09df0cfc856c results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto dataset_prepared_path: null datasets: - data_files: - a7412d8c8f805ddf_train_data.json ds_type: json format: custom path: a7412d8c8f805ddf_train_data.json type: field: null field_input: null field_instruction: premise field_output: hypothesis field_system: null format: null no_input_format: null system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 2 flash_attention: null fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: taopanda-2/a31b520a-ea45-423d-a41b-09df0cfc856c learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 32 lora_target_linear: true lr_scheduler: cosine micro_batch_size: 2 model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: ./outputs/lora-out/taopanda-2_ba28db81-f399-44e5-bdef-7af8dcf5a4ca pad_to_sequence_len: null resume_from_checkpoint: null sample_packing: false saves_per_epoch: 1 seed: 91813 sequence_len: 2048 special_tokens: null strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: fatcat87-taopanda wandb_log_model: null wandb_mode: online wandb_name: taopanda-2_ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_project: subnet56 wandb_runid: taopanda-2_ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_watch: null warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/fatcat87-taopanda/subnet56/runs/sk5fvbjg) # a31b520a-ea45-423d-a41b-09df0cfc856c This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.9854 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 91813 - distributed_type: multi-GPU - num_devices: 4 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - total_eval_batch_size: 8 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - num_epochs: 1 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 4.4358 \| 0.0006 \| 1 \| 4.7686 \| \| 0.8153 \| 0.4998 \| 823 \| 1.0209 \| \| 0.9222 \| 0.9997 \| 1646 \| 0.9854 \| ### Framework versions - PEFT 0.11.1 - Transformers 4.42.3 - Pytorch 2.3.0+cu121 - Datasets 2.19.1 - Tokenizers 0.19.1
nat-hunt/13925d41-82bd-407e-b7bd-93b7227d91e6	nat-hunt	2025-01-20T23:49:00Z	8	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:unsloth/Meta-Llama-3.1-8B", "base_model:adapter:unsloth/Meta-Llama-3.1-8B", "license:llama3.1", "region:us" ]	null	2025-01-20T23:45:49Z	--- library_name: peft license: llama3.1 base_model: unsloth/Meta-Llama-3.1-8B tags: - axolotl - generated_from_trainer model-index: - name: 13925d41-82bd-407e-b7bd-93b7227d91e6 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Meta-Llama-3.1-8B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 91e193d3dca1611f_train_data.json ds_type: json format: custom path: /workspace/input_data/91e193d3dca1611f_train_data.json type: field_input: parent_id field_instruction: role field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: nat-hunt/13925d41-82bd-407e-b7bd-93b7227d91e6 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/91e193d3dca1611f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 856a9aac-189f-40f7-b27c-c5616995b0d1 wandb_project: Birthday-SN56-25-Gradients-On-Demand wandb_run: your_name wandb_runid: 856a9aac-189f-40f7-b27c-c5616995b0d1 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 13925d41-82bd-407e-b7bd-93b7227d91e6 This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.0004 \| 1 \| nan \| \| 0.0 \| 0.0011 \| 3 \| nan \| \| 0.0 \| 0.0021 \| 6 \| nan \| \| 0.0 \| 0.0032 \| 9 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
LHRuig/doubleexpo	LHRuig	2025-01-20T23:48:34Z	9	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:48:28Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # doubleexpo <Gallery /> ## Model description doubleexpo lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/doubleexpo/tree/main) them in the Files & versions tab.
mradermacher/QwQ-56B-Ghost-i1-GGUF	mradermacher	2025-01-20T23:48:26Z	513	0	transformers	[ "transformers", "gguf", "mergekit", "merge", "uncensored", "abliterated", "chat", "en", "base_model:JackCloudman/QwQ-56B-Ghost", "base_model:quantized:JackCloudman/QwQ-56B-Ghost", "license:apache-2.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	null	2025-01-20T12:39:13Z	--- base_model: JackCloudman/QwQ-56B-Ghost language: - en library_name: transformers license: apache-2.0 quantized_by: mradermacher tags: - mergekit - merge - uncensored - abliterated - chat --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: nicoboss --> weighted/imatrix quants of https://huggingface.co/JackCloudman/QwQ-56B-Ghost <!-- provided-files --> static quants are available at https://huggingface.co/mradermacher/QwQ-56B-Ghost-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ1_S.gguf) \| i1-IQ1_S \| 12.2 \| for the desperate \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ1_M.gguf) \| i1-IQ1_M \| 13.4 \| mostly desperate \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ2_XXS.gguf) \| i1-IQ2_XXS \| 15.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ2_XS.gguf) \| i1-IQ2_XS \| 16.9 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ2_S.gguf) \| i1-IQ2_S \| 17.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ2_M.gguf) \| i1-IQ2_M \| 19.2 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q2_K_S.gguf) \| i1-Q2_K_S \| 19.5 \| very low quality \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q2_K.gguf) \| i1-Q2_K \| 21.0 \| IQ3_XXS probably better \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ3_XXS.gguf) \| i1-IQ3_XXS \| 21.9 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ3_XS.gguf) \| i1-IQ3_XS \| 23.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q3_K_S.gguf) \| i1-Q3_K_S \| 24.6 \| IQ3_XS probably better \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ3_S.gguf) \| i1-IQ3_S \| 24.6 \| beats Q3_K* \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ3_M.gguf) \| i1-IQ3_M \| 25.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q3_K_M.gguf) \| i1-Q3_K_M \| 27.3 \| IQ3_S probably better \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q3_K_L.gguf) \| i1-Q3_K_L \| 29.5 \| IQ3_M probably better \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ4_XS.gguf) \| i1-IQ4_XS \| 30.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q4_0.gguf) \| i1-Q4_0 \| 32.0 \| fast, low quality \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q4_K_S.gguf) \| i1-Q4_K_S \| 32.2 \| optimal size/speed/quality \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q4_K_M.gguf) \| i1-Q4_K_M \| 34.0 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q4_1.gguf) \| i1-Q4_1 \| 35.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q5_K_S.gguf) \| i1-Q5_K_S \| 38.8 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q5_K_M.gguf) \| i1-Q5_K_M \| 39.9 \| \| \| [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q6_K.gguf) \| i1-Q6_K \| 46.2 \| practically like static Q6_K \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to. <!-- end -->
mradermacher/LwQ-Reasoner-10B-i1-GGUF	mradermacher	2025-01-20T23:48:26Z	580	0	transformers	[ "transformers", "gguf", "LlamaWithQuestions", "CoT", "Reasoner", "LWQ", "en", "base_model:prithivMLmods/LwQ-Reasoner-10B", "base_model:quantized:prithivMLmods/LwQ-Reasoner-10B", "license:llama3.1", "endpoints_compatible", "region:us", "imatrix" ]	null	2025-01-20T22:25:16Z	--- base_model: prithivMLmods/LwQ-Reasoner-10B language: - en library_name: transformers license: llama3.1 quantized_by: mradermacher tags: - LlamaWithQuestions - CoT - Reasoner - LWQ --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: nicoboss --> weighted/imatrix quants of https://huggingface.co/prithivMLmods/LwQ-Reasoner-10B <!-- provided-files --> static quants are available at https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ1_S.gguf) \| i1-IQ1_S \| 2.5 \| for the desperate \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ1_M.gguf) \| i1-IQ1_M \| 2.7 \| mostly desperate \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ2_XXS.gguf) \| i1-IQ2_XXS \| 3.0 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ2_XS.gguf) \| i1-IQ2_XS \| 3.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ2_S.gguf) \| i1-IQ2_S \| 3.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ2_M.gguf) \| i1-IQ2_M \| 3.7 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q2_K_S.gguf) \| i1-Q2_K_S \| 3.8 \| very low quality \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q2_K.gguf) \| i1-Q2_K \| 4.0 \| IQ3_XXS probably better \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ3_XXS.gguf) \| i1-IQ3_XXS \| 4.2 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ3_XS.gguf) \| i1-IQ3_XS \| 4.5 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ3_S.gguf) \| i1-IQ3_S \| 4.7 \| beats Q3_K* \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q3_K_S.gguf) \| i1-Q3_K_S \| 4.7 \| IQ3_XS probably better \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ3_M.gguf) \| i1-IQ3_M \| 4.8 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q3_K_M.gguf) \| i1-Q3_K_M \| 5.2 \| IQ3_S probably better \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q3_K_L.gguf) \| i1-Q3_K_L \| 5.6 \| IQ3_M probably better \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ4_XS.gguf) \| i1-IQ4_XS \| 5.7 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ4_NL.gguf) \| i1-IQ4_NL \| 6.0 \| prefer IQ4_XS \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q4_0.gguf) \| i1-Q4_0 \| 6.0 \| fast, low quality \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q4_K_S.gguf) \| i1-Q4_K_S \| 6.1 \| optimal size/speed/quality \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q4_K_M.gguf) \| i1-Q4_K_M \| 6.4 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q4_1.gguf) \| i1-Q4_1 \| 6.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q5_K_S.gguf) \| i1-Q5_K_S \| 7.2 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q5_K_M.gguf) \| i1-Q5_K_M \| 7.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q6_K.gguf) \| i1-Q6_K \| 8.6 \| practically like static Q6_K \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to. <!-- end -->
jncraton/DeepSeek-R1-Distill-Llama-8B-ct2-int8	jncraton	2025-01-20T23:47:44Z	12	0	null	[ "region:us" ]	null	2025-01-20T23:33:07Z	# DeepSeek-R1 <!-- markdownlint-disable first-line-h1 --> <!-- markdownlint-disable html --> <!-- markdownlint-disable no-duplicate-header --> <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- Post-Training: Large-Scale Reinforcement Learning on the Base Model - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- Distillation: Smaller Models Can Be Powerful Too - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> \| Model \| #Total Params \| #Activated Params \| Context Length \| Download \| \| :------------: \| :------------: \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Zero \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) \| \| DeepSeek-R1 \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) \| </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> \| Model \| Base Model \| Download \| \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Distill-Qwen-1.5B \| [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) \| \| DeepSeek-R1-Distill-Qwen-7B \| [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) \| \| DeepSeek-R1-Distill-Llama-8B \| [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) \| \| DeepSeek-R1-Distill-Qwen-14B \| [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) \| \|DeepSeek-R1-Distill-Qwen-32B \| [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) \| \| DeepSeek-R1-Distill-Llama-70B \| [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) \| </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> \| Category \| Benchmark (Metric) \| Claude-3.5-Sonnet-1022 \| GPT-4o 0513 \| DeepSeek V3 \| OpenAI o1-mini \| OpenAI o1-1217 \| DeepSeek R1 \| \|----------\|-------------------\|----------------------\|------------\|--------------\|----------------\|------------\|--------------\| \| \| Architecture \| - \| - \| MoE \| - \| - \| MoE \| \| \| # Activated Params \| - \| - \| 37B \| - \| - \| 37B \| \| \| # Total Params \| - \| - \| 671B \| - \| - \| 671B \| \| English \| MMLU (Pass@1) \| 88.3 \| 87.2 \| 88.5 \| 85.2 \| 91.8 \| 90.8 \| \| \| MMLU-Redux (EM) \| 88.9 \| 88.0 \| 89.1 \| 86.7 \| - \| 92.9 \| \| \| MMLU-Pro (EM) \| 78.0 \| 72.6 \| 75.9 \| 80.3 \| - \| 84.0 \| \| \| DROP (3-shot F1) \| 88.3 \| 83.7 \| 91.6 \| 83.9 \| 90.2 \| 92.2 \| \| \| IF-Eval (Prompt Strict) \| 86.5 \| 84.3 \| 86.1 \| 84.8 \| - \| 83.3 \| \| \| GPQA-Diamond (Pass@1) \| 65.0 \| 49.9 \| 59.1 \| 60.0 \| 75.7 \| 71.5 \| \| \| SimpleQA (Correct) \| 28.4 \| 38.2 \| 24.9 \| 7.0 \| 47.0 \| 30.1 \| \| \| FRAMES (Acc.) \| 72.5 \| 80.5 \| 73.3 \| 76.9 \| - \| 82.5 \| \| \| AlpacaEval2.0 (LC-winrate) \| 52.0 \| 51.1 \| 70.0 \| 57.8 \| - \| 87.6 \| \| \| ArenaHard (GPT-4-1106) \| 85.2 \| 80.4 \| 85.5 \| 92.0 \| - \| 92.3 \| \| Code \| LiveCodeBench (Pass@1-COT) \| 33.8 \| 34.2 \| - \| 53.8 \| 63.4 \| 65.9 \| \| \| Codeforces (Percentile) \| 20.3 \| 23.6 \| 58.7 \| 93.4 \| 96.6 \| 96.3 \| \| \| Codeforces (Rating) \| 717 \| 759 \| 1134 \| 1820 \| 2061 \| 2029 \| \| \| SWE Verified (Resolved) \| 50.8 \| 38.8 \| 42.0 \| 41.6 \| 48.9 \| 49.2 \| \| \| Aider-Polyglot (Acc.) \| 45.3 \| 16.0 \| 49.6 \| 32.9 \| 61.7 \| 53.3 \| \| Math \| AIME 2024 (Pass@1) \| 16.0 \| 9.3 \| 39.2 \| 63.6 \| 79.2 \| 79.8 \| \| \| MATH-500 (Pass@1) \| 78.3 \| 74.6 \| 90.2 \| 90.0 \| 96.4 \| 97.3 \| \| \| CNMO 2024 (Pass@1) \| 13.1 \| 10.8 \| 43.2 \| 67.6 \| - \| 78.8 \| \| Chinese \| CLUEWSC (EM) \| 85.4 \| 87.9 \| 90.9 \| 89.9 \| - \| 92.8 \| \| \| C-Eval (EM) \| 76.7 \| 76.0 \| 86.5 \| 68.9 \| - \| 91.8 \| \| \| C-SimpleQA (Correct) \| 55.4 \| 58.7 \| 68.0 \| 40.3 \| - \| 63.7 \| </div> ### Distilled Model Evaluation <div align="center"> \| Model \| AIME 2024 pass@1 \| AIME 2024 cons@64 \| MATH-500 pass@1 \| GPQA Diamond pass@1 \| LiveCodeBench pass@1 \| CodeForces rating \| \|------------------------------------------\|------------------\|-------------------\|-----------------\|----------------------\|----------------------\|-------------------\| \| GPT-4o-0513 \| 9.3 \| 13.4 \| 74.6 \| 49.9 \| 32.9 \| 759 \| \| Claude-3.5-Sonnet-1022 \| 16.0 \| 26.7 \| 78.3 \| 65.0 \| 38.9 \| 717 \| \| o1-mini \| 63.6 \| 80.0 \| 90.0 \| 60.0 \| 53.8 \| 1820 \| \| QwQ-32B-Preview \| 44.0 \| 60.0 \| 90.6 \| 54.5 \| 41.9 \| 1316 \| \| DeepSeek-R1-Distill-Qwen-1.5B \| 28.9 \| 52.7 \| 83.9 \| 33.8 \| 16.9 \| 954 \| \| DeepSeek-R1-Distill-Qwen-7B \| 55.5 \| 83.3 \| 92.8 \| 49.1 \| 37.6 \| 1189 \| \| DeepSeek-R1-Distill-Qwen-14B \| 69.7 \| 80.0 \| 93.9 \| 59.1 \| 53.1 \| 1481 \| \| DeepSeek-R1-Distill-Qwen-32B \| 72.6 \| 83.3 \| 94.3 \| 62.1 \| 57.2 \| 1691 \| \| DeepSeek-R1-Distill-Llama-8B \| 50.4 \| 80.0 \| 89.1 \| 49.0 \| 39.6 \| 1205 \| \| DeepSeek-R1-Distill-Llama-70B \| 70.0 \| 86.7 \| 94.5 \| 65.2 \| 57.5 \| 1633 \| </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output. ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).
Sakalti/SJT-4.5B	Sakalti	2025-01-20T23:47:23Z	152	0	transformers	[ "transformers", "safetensors", "qwen2", "text-generation", "text-generation-inference", "unsloth", "trl", "conversational", "en", "ja", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "4-bit", "bitsandbytes", "region:us" ]	text-generation	2024-12-23T08:33:51Z	--- base_model: unsloth/qwen2.5-7b-instruct-bnb-4bit inference: true tags: - text-generation-inference - transformers - unsloth - qwen2 - trl license: apache-2.0 language: - en - ja widget: - messages: - role: user content: こんにちは！ - messages: - role: user content: 魚を捌くのは難しいですか？ - messages: - role: user content: 日本の首都はどこですか？ - messages: - role: user content: hello! - messages: - role: user content: こんにちは！ - messages: - role: user content: whats is the capital of japan? - messages: - role: user content: Who are you? - messages: - role: user content: 你好 --- # Uploaded model - Developed by: Sakalti - License: apache-2.0 - Finetuned from model : unsloth/qwen2.5-7b-instruct-bnb-4bit This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
JacksonBrune/bf94c46f-78a2-4147-b4e1-1ca46e6cb0ef	JacksonBrune	2025-01-20T23:45:12Z	6	0	peft	[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "base_model:unsloth/Phi-3-medium-4k-instruct", "base_model:adapter:unsloth/Phi-3-medium-4k-instruct", "license:mit", "region:us" ]	null	2025-01-20T22:29:38Z	--- library_name: peft license: mit base_model: unsloth/Phi-3-medium-4k-instruct tags: - axolotl - generated_from_trainer model-index: - name: bf94c46f-78a2-4147-b4e1-1ca46e6cb0ef results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Phi-3-medium-4k-instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 41d403c8b37c92fc_train_data.json ds_type: json format: custom path: /workspace/input_data/41d403c8b37c92fc_train_data.json type: field_input: mesh_terms field_instruction: title field_output: abstract format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: JacksonBrune/bf94c46f-78a2-4147-b4e1-1ca46e6cb0ef hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/41d403c8b37c92fc_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075ea541-bd04-429e-a989-c49dabc36fc3 wandb_project: birthdya-sn56-18-Gradients-On-Demand wandb_run: your_name wandb_runid: 075ea541-bd04-429e-a989-c49dabc36fc3 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # bf94c46f-78a2-4147-b4e1-1ca46e6cb0ef This model is a fine-tuned version of [unsloth/Phi-3-medium-4k-instruct](https://huggingface.co/unsloth/Phi-3-medium-4k-instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.0000 \| 1 \| nan \| \| 0.0 \| 0.0001 \| 3 \| nan \| \| 0.0 \| 0.0001 \| 6 \| nan \| \| 0.0 \| 0.0002 \| 9 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
marialvsantiago/7b9a7390-1246-446a-87dd-03f5c20759c1	marialvsantiago	2025-01-20T23:45:06Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:unsloth/SmolLM-1.7B", "base_model:adapter:unsloth/SmolLM-1.7B", "license:apache-2.0", "region:us" ]	null	2025-01-20T23:15:21Z	--- library_name: peft license: apache-2.0 base_model: unsloth/SmolLM-1.7B tags: - axolotl - generated_from_trainer model-index: - name: 7b9a7390-1246-446a-87dd-03f5c20759c1 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/SmolLM-1.7B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - b50bb242ea24ad3f_train_data.json ds_type: json format: custom path: /workspace/input_data/b50bb242ea24ad3f_train_data.json type: field_input: my_solu field_instruction: prompt field_output: solution format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: marialvsantiago/7b9a7390-1246-446a-87dd-03f5c20759c1 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 78GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/b50bb242ea24ad3f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 8caf838c-eaff-45bc-b751-9573db70c518 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 8caf838c-eaff-45bc-b751-9573db70c518 warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # 7b9a7390-1246-446a-87dd-03f5c20759c1 This model is a fine-tuned version of [unsloth/SmolLM-1.7B](https://huggingface.co/unsloth/SmolLM-1.7B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0001 \| 1 \| nan \| \| 0.0 \| 0.0005 \| 5 \| nan \| \| 0.0 \| 0.0010 \| 10 \| nan \| \| 0.0 \| 0.0015 \| 15 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
mmnga/DeepSeek-R1-Distill-Qwen-32B-gguf	mmnga	2025-01-20T23:44:54Z	3,380	3	null	[ "gguf", "qwen2", "en", "ja", "dataset:TFMC/imatrix-dataset-for-japanese-llm", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", "license:mit", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	null	2025-01-20T16:54:59Z	--- license: mit language: - en - ja datasets: - TFMC/imatrix-dataset-for-japanese-llm tags: - qwen2 base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --- # DeepSeek-R1-Distill-Qwen-32B-gguf [deepseek-aiさんが公開しているDeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B)のggufフォーマット変換版です。 imatrixのデータは[TFMC/imatrix-dataset-for-japanese-llm](https://huggingface.co/datasets/TFMC/imatrix-dataset-for-japanese-llm)を使用して作成しました。 ## Usage ``` git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build -DGGML_CUDA=ON cmake --build build --config Release build/bin/llama-cli -m 'DeepSeek-R1-Distill-Qwen-32B-gguf' -n 128 -c 128 -p 'あなたはプロの料理人です。レシピを教えて' -cnv ```
Tejasvisudugureddy/distilled_dpo_chatbot	Tejasvisudugureddy	2025-01-20T23:44:50Z	213	0	transformers	[ "transformers", "tensorboard", "safetensors", "llama", "text-generation", "generated_from_trainer", "conversational", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-01-20T23:44:34Z	--- library_name: transformers tags: - generated_from_trainer model-index: - name: distilled_dpo_chatbot results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # distilled_dpo_chatbot This model was trained from scratch on an unknown dataset. ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 3.0 ### Framework versions - Transformers 4.47.0 - Pytorch 2.5.1+cu121 - Datasets 3.2.0 - Tokenizers 0.21.0
LHRuig/proultra	LHRuig	2025-01-20T23:44:49Z	9	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:44:29Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # proultra <Gallery /> ## Model description proultra lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/proultra/tree/main) them in the Files & versions tab.
LHRuig/reallora	LHRuig	2025-01-20T23:41:59Z	7	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:41:55Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # reallora <Gallery /> ## Model description reallora lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/reallora/tree/main) them in the Files & versions tab.
LHRuig/db0real	LHRuig	2025-01-20T23:41:19Z	16	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:41:17Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # db0real <Gallery /> ## Model description db0real lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/db0real/tree/main) them in the Files & versions tab.
MayBashendy/ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k19_task5_organization	MayBashendy	2025-01-20T23:41:10Z	5	0	transformers	[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-classification	2025-01-20T23:22:32Z	--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k19_task5_organization results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k19_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.3225 - Qwk: 0.0 - Mse: 1.3225 - Rmse: 1.1500 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Qwk \| Mse \| Rmse \| \|:-------------:\|:-------:\|:----:\|:---------------:\|:-------:\|:------:\|:------:\| \| No log \| 0.0435 \| 2 \| 3.9365 \| -0.0232 \| 3.9365 \| 1.9841 \| \| No log \| 0.0870 \| 4 \| 2.7213 \| -0.0305 \| 2.7213 \| 1.6496 \| \| No log \| 0.1304 \| 6 \| 2.3416 \| -0.0372 \| 2.3416 \| 1.5302 \| \| No log \| 0.1739 \| 8 \| 1.4991 \| 0.0854 \| 1.4991 \| 1.2244 \| \| No log \| 0.2174 \| 10 \| 1.1113 \| 0.2268 \| 1.1113 \| 1.0542 \| \| No log \| 0.2609 \| 12 \| 1.1950 \| 0.0731 \| 1.1950 \| 1.0932 \| \| No log \| 0.3043 \| 14 \| 1.1082 \| 0.2094 \| 1.1082 \| 1.0527 \| \| No log \| 0.3478 \| 16 \| 1.2264 \| 0.0884 \| 1.2264 \| 1.1074 \| \| No log \| 0.3913 \| 18 \| 1.6641 \| -0.0661 \| 1.6641 \| 1.2900 \| \| No log \| 0.4348 \| 20 \| 1.5899 \| -0.1122 \| 1.5899 \| 1.2609 \| \| No log \| 0.4783 \| 22 \| 1.3669 \| 0.0519 \| 1.3669 \| 1.1692 \| \| No log \| 0.5217 \| 24 \| 1.3565 \| -0.0417 \| 1.3565 \| 1.1647 \| \| No log \| 0.5652 \| 26 \| 1.3698 \| -0.0032 \| 1.3698 \| 1.1704 \| \| No log \| 0.6087 \| 28 \| 1.3109 \| -0.0497 \| 1.3109 \| 1.1450 \| \| No log \| 0.6522 \| 30 \| 1.0949 \| 0.1589 \| 1.0949 \| 1.0464 \| \| No log \| 0.6957 \| 32 \| 1.0503 \| 0.2114 \| 1.0503 \| 1.0248 \| \| No log \| 0.7391 \| 34 \| 1.1274 \| 0.1160 \| 1.1274 \| 1.0618 \| \| No log \| 0.7826 \| 36 \| 1.4414 \| -0.0881 \| 1.4414 \| 1.2006 \| \| No log \| 0.8261 \| 38 \| 1.6338 \| -0.1445 \| 1.6338 \| 1.2782 \| \| No log \| 0.8696 \| 40 \| 1.2646 \| 0.0622 \| 1.2646 \| 1.1245 \| \| No log \| 0.9130 \| 42 \| 1.0736 \| 0.1864 \| 1.0736 \| 1.0361 \| \| No log \| 0.9565 \| 44 \| 1.0625 \| 0.2911 \| 1.0625 \| 1.0308 \| \| No log \| 1.0 \| 46 \| 1.1263 \| 0.2647 \| 1.1263 \| 1.0613 \| \| No log \| 1.0435 \| 48 \| 1.3547 \| -0.0249 \| 1.3547 \| 1.1639 \| \| No log \| 1.0870 \| 50 \| 1.6222 \| -0.1135 \| 1.6222 \| 1.2737 \| \| No log \| 1.1304 \| 52 \| 1.4905 \| -0.0623 \| 1.4905 \| 1.2209 \| \| No log \| 1.1739 \| 54 \| 1.2550 \| 0.0911 \| 1.2550 \| 1.1203 \| \| No log \| 1.2174 \| 56 \| 1.0641 \| 0.1857 \| 1.0641 \| 1.0315 \| \| No log \| 1.2609 \| 58 \| 1.0561 \| 0.2179 \| 1.0561 \| 1.0277 \| \| No log \| 1.3043 \| 60 \| 1.1071 \| 0.1761 \| 1.1071 \| 1.0522 \| \| No log \| 1.3478 \| 62 \| 1.2148 \| 0.2021 \| 1.2148 \| 1.1022 \| \| No log \| 1.3913 \| 64 \| 1.3043 \| 0.1489 \| 1.3043 \| 1.1421 \| \| No log \| 1.4348 \| 66 \| 1.2530 \| 0.1609 \| 1.2530 \| 1.1194 \| \| No log \| 1.4783 \| 68 \| 1.1703 \| 0.2312 \| 1.1703 \| 1.0818 \| \| No log \| 1.5217 \| 70 \| 1.1921 \| 0.1801 \| 1.1921 \| 1.0918 \| \| No log \| 1.5652 \| 72 \| 1.2234 \| 0.1370 \| 1.2234 \| 1.1061 \| \| No log \| 1.6087 \| 74 \| 1.4115 \| 0.0939 \| 1.4115 \| 1.1881 \| \| No log \| 1.6522 \| 76 \| 1.5486 \| 0.1111 \| 1.5486 \| 1.2444 \| \| No log \| 1.6957 \| 78 \| 1.4303 \| 0.0861 \| 1.4303 \| 1.1959 \| \| No log \| 1.7391 \| 80 \| 1.2314 \| 0.1500 \| 1.2314 \| 1.1097 \| \| No log \| 1.7826 \| 82 \| 1.0416 \| 0.2226 \| 1.0416 \| 1.0206 \| \| No log \| 1.8261 \| 84 \| 0.9985 \| 0.2818 \| 0.9985 \| 0.9993 \| \| No log \| 1.8696 \| 86 \| 1.0136 \| 0.2251 \| 1.0136 \| 1.0068 \| \| No log \| 1.9130 \| 88 \| 1.1889 \| 0.0961 \| 1.1889 \| 1.0904 \| \| No log \| 1.9565 \| 90 \| 1.3523 \| 0.0541 \| 1.3523 \| 1.1629 \| \| No log \| 2.0 \| 92 \| 1.3806 \| 0.0541 \| 1.3806 \| 1.1750 \| \| No log \| 2.0435 \| 94 \| 1.2957 \| 0.1170 \| 1.2957 \| 1.1383 \| \| No log \| 2.0870 \| 96 \| 1.2446 \| 0.2577 \| 1.2446 \| 1.1156 \| \| No log \| 2.1304 \| 98 \| 1.2931 \| 0.2934 \| 1.2931 \| 1.1371 \| \| No log \| 2.1739 \| 100 \| 1.3752 \| 0.2647 \| 1.3752 \| 1.1727 \| \| No log \| 2.2174 \| 102 \| 1.5603 \| 0.2239 \| 1.5603 \| 1.2491 \| \| No log \| 2.2609 \| 104 \| 1.5673 \| 0.1943 \| 1.5673 \| 1.2519 \| \| No log \| 2.3043 \| 106 \| 1.3581 \| 0.1880 \| 1.3581 \| 1.1654 \| \| No log \| 2.3478 \| 108 \| 1.2639 \| 0.2203 \| 1.2639 \| 1.1242 \| \| No log \| 2.3913 \| 110 \| 1.3721 \| 0.1744 \| 1.3721 \| 1.1713 \| \| No log \| 2.4348 \| 112 \| 1.3828 \| 0.1744 \| 1.3828 \| 1.1759 \| \| No log \| 2.4783 \| 114 \| 1.3706 \| 0.0931 \| 1.3706 \| 1.1707 \| \| No log \| 2.5217 \| 116 \| 1.3150 \| 0.0278 \| 1.3150 \| 1.1467 \| \| No log \| 2.5652 \| 118 \| 1.3244 \| 0.0278 \| 1.3244 \| 1.1508 \| \| No log \| 2.6087 \| 120 \| 1.3610 \| 0.0571 \| 1.3610 \| 1.1666 \| \| No log \| 2.6522 \| 122 \| 1.4379 \| 0.1438 \| 1.4379 \| 1.1991 \| \| No log \| 2.6957 \| 124 \| 1.5829 \| 0.2084 \| 1.5829 \| 1.2581 \| \| No log \| 2.7391 \| 126 \| 1.6933 \| 0.1630 \| 1.6933 \| 1.3013 \| \| No log \| 2.7826 \| 128 \| 1.7564 \| 0.1565 \| 1.7564 \| 1.3253 \| \| No log \| 2.8261 \| 130 \| 1.6753 \| 0.1531 \| 1.6753 \| 1.2943 \| \| No log \| 2.8696 \| 132 \| 1.5731 \| 0.2084 \| 1.5731 \| 1.2542 \| \| No log \| 2.9130 \| 134 \| 1.3438 \| 0.1935 \| 1.3438 \| 1.1592 \| \| No log \| 2.9565 \| 136 \| 1.1707 \| 0.3322 \| 1.1707 \| 1.0820 \| \| No log \| 3.0 \| 138 \| 1.1864 \| 0.2986 \| 1.1864 \| 1.0892 \| \| No log \| 3.0435 \| 140 \| 1.3843 \| 0.1727 \| 1.3843 \| 1.1765 \| \| No log \| 3.0870 \| 142 \| 1.3959 \| 0.2026 \| 1.3959 \| 1.1815 \| \| No log \| 3.1304 \| 144 \| 1.2520 \| 0.1750 \| 1.2520 \| 1.1189 \| \| No log \| 3.1739 \| 146 \| 1.0363 \| 0.2836 \| 1.0363 \| 1.0180 \| \| No log \| 3.2174 \| 148 \| 1.0039 \| 0.2814 \| 1.0039 \| 1.0020 \| \| No log \| 3.2609 \| 150 \| 1.0200 \| 0.2747 \| 1.0200 \| 1.0099 \| \| No log \| 3.3043 \| 152 \| 1.1611 \| 0.1751 \| 1.1611 \| 1.0775 \| \| No log \| 3.3478 \| 154 \| 1.2712 \| 0.2241 \| 1.2712 \| 1.1275 \| \| No log \| 3.3913 \| 156 \| 1.3275 \| 0.2026 \| 1.3275 \| 1.1522 \| \| No log \| 3.4348 \| 158 \| 1.3712 \| 0.2015 \| 1.3712 \| 1.1710 \| \| No log \| 3.4783 \| 160 \| 1.4292 \| 0.2126 \| 1.4292 \| 1.1955 \| \| No log \| 3.5217 \| 162 \| 1.3521 \| 0.1744 \| 1.3521 \| 1.1628 \| \| No log \| 3.5652 \| 164 \| 1.2403 \| 0.1052 \| 1.2403 \| 1.1137 \| \| No log \| 3.6087 \| 166 \| 1.2213 \| 0.1052 \| 1.2213 \| 1.1051 \| \| No log \| 3.6522 \| 168 \| 1.2553 \| 0.1407 \| 1.2553 \| 1.1204 \| \| No log \| 3.6957 \| 170 \| 1.3345 \| 0.1744 \| 1.3345 \| 1.1552 \| \| No log \| 3.7391 \| 172 \| 1.3148 \| 0.1407 \| 1.3148 \| 1.1467 \| \| No log \| 3.7826 \| 174 \| 1.2434 \| 0.1407 \| 1.2434 \| 1.1151 \| \| No log \| 3.8261 \| 176 \| 1.1953 \| 0.1744 \| 1.1953 \| 1.0933 \| \| No log \| 3.8696 \| 178 \| 1.1782 \| 0.1838 \| 1.1782 \| 1.0855 \| \| No log \| 3.9130 \| 180 \| 1.2119 \| 0.2372 \| 1.2119 \| 1.1009 \| \| No log \| 3.9565 \| 182 \| 1.2574 \| 0.1952 \| 1.2574 \| 1.1213 \| \| No log \| 4.0 \| 184 \| 1.2950 \| 0.1288 \| 1.2950 \| 1.1380 \| \| No log \| 4.0435 \| 186 \| 1.2514 \| 0.1952 \| 1.2514 \| 1.1187 \| \| No log \| 4.0870 \| 188 \| 1.0793 \| 0.1628 \| 1.0793 \| 1.0389 \| \| No log \| 4.1304 \| 190 \| 0.9864 \| 0.0990 \| 0.9864 \| 0.9932 \| \| No log \| 4.1739 \| 192 \| 1.0281 \| 0.1654 \| 1.0281 \| 1.0139 \| \| No log \| 4.2174 \| 194 \| 1.2053 \| 0.1943 \| 1.2053 \| 1.0979 \| \| No log \| 4.2609 \| 196 \| 1.4822 \| 0.2733 \| 1.4822 \| 1.2174 \| \| No log \| 4.3043 \| 198 \| 1.6190 \| 0.2733 \| 1.6190 \| 1.2724 \| \| No log \| 4.3478 \| 200 \| 1.5802 \| 0.2568 \| 1.5802 \| 1.2571 \| \| No log \| 4.3913 \| 202 \| 1.4305 \| 0.2089 \| 1.4305 \| 1.1961 \| \| No log \| 4.4348 \| 204 \| 1.2446 \| 0.1654 \| 1.2446 \| 1.1156 \| \| No log \| 4.4783 \| 206 \| 1.1579 \| 0.1226 \| 1.1579 \| 1.0761 \| \| No log \| 4.5217 \| 208 \| 1.2399 \| -0.0929 \| 1.2399 \| 1.1135 \| \| No log \| 4.5652 \| 210 \| 1.3674 \| -0.0263 \| 1.3674 \| 1.1693 \| \| No log \| 4.6087 \| 212 \| 1.4571 \| 0.0935 \| 1.4571 \| 1.2071 \| \| No log \| 4.6522 \| 214 \| 1.6547 \| 0.1565 \| 1.6547 \| 1.2864 \| \| No log \| 4.6957 \| 216 \| 1.7420 \| 0.1476 \| 1.7420 \| 1.3198 \| \| No log \| 4.7391 \| 218 \| 1.6654 \| 0.1423 \| 1.6654 \| 1.2905 \| \| No log \| 4.7826 \| 220 \| 1.4261 \| 0.2062 \| 1.4261 \| 1.1942 \| \| No log \| 4.8261 \| 222 \| 1.1989 \| 0.2038 \| 1.1989 \| 1.0950 \| \| No log \| 4.8696 \| 224 \| 1.1364 \| 0.1976 \| 1.1364 \| 1.0660 \| \| No log \| 4.9130 \| 226 \| 1.1866 \| 0.2126 \| 1.1866 \| 1.0893 \| \| No log \| 4.9565 \| 228 \| 1.3037 \| 0.2126 \| 1.3037 \| 1.1418 \| \| No log \| 5.0 \| 230 \| 1.4564 \| 0.2342 \| 1.4564 \| 1.2068 \| \| No log \| 5.0435 \| 232 \| 1.5256 \| 0.2694 \| 1.5256 \| 1.2352 \| \| No log \| 5.0870 \| 234 \| 1.5762 \| 0.2940 \| 1.5762 \| 1.2555 \| \| No log \| 5.1304 \| 236 \| 1.5906 \| 0.2940 \| 1.5906 \| 1.2612 \| \| No log \| 5.1739 \| 238 \| 1.6182 \| 0.2974 \| 1.6182 \| 1.2721 \| \| No log \| 5.2174 \| 240 \| 1.6432 \| 0.2974 \| 1.6432 \| 1.2819 \| \| No log \| 5.2609 \| 242 \| 1.5135 \| 0.2694 \| 1.5135 \| 1.2302 \| \| No log \| 5.3043 \| 244 \| 1.3818 \| 0.2170 \| 1.3818 \| 1.1755 \| \| No log \| 5.3478 \| 246 \| 1.2951 \| 0.2123 \| 1.2951 \| 1.1380 \| \| No log \| 5.3913 \| 248 \| 1.2601 \| 0.1850 \| 1.2601 \| 1.1225 \| \| No log \| 5.4348 \| 250 \| 1.2575 \| 0.2170 \| 1.2574 \| 1.1214 \| \| No log \| 5.4783 \| 252 \| 1.2783 \| 0.2117 \| 1.2783 \| 1.1306 \| \| No log \| 5.5217 \| 254 \| 1.3535 \| 0.2117 \| 1.3535 \| 1.1634 \| \| No log \| 5.5652 \| 256 \| 1.4060 \| 0.2391 \| 1.4060 \| 1.1857 \| \| No log \| 5.6087 \| 258 \| 1.3457 \| 0.2117 \| 1.3457 \| 1.1600 \| \| No log \| 5.6522 \| 260 \| 1.2294 \| 0.2170 \| 1.2294 \| 1.1088 \| \| No log \| 5.6957 \| 262 \| 1.2760 \| 0.2170 \| 1.2760 \| 1.1296 \| \| No log \| 5.7391 \| 264 \| 1.4787 \| 0.2653 \| 1.4787 \| 1.2160 \| \| No log \| 5.7826 \| 266 \| 1.5640 \| 0.2391 \| 1.5640 \| 1.2506 \| \| No log \| 5.8261 \| 268 \| 1.5706 \| 0.2062 \| 1.5706 \| 1.2532 \| \| No log \| 5.8696 \| 270 \| 1.4463 \| 0.2239 \| 1.4463 \| 1.2026 \| \| No log \| 5.9130 \| 272 \| 1.2451 \| 0.1407 \| 1.2451 \| 1.1158 \| \| No log \| 5.9565 \| 274 \| 1.1629 \| 0.0401 \| 1.1629 \| 1.0784 \| \| No log \| 6.0 \| 276 \| 1.1448 \| 0.0401 \| 1.1448 \| 1.0699 \| \| No log \| 6.0435 \| 278 \| 1.1363 \| 0.0 \| 1.1363 \| 1.0660 \| \| No log \| 6.0870 \| 280 \| 1.2335 \| 0.0781 \| 1.2335 \| 1.1106 \| \| No log \| 6.1304 \| 282 \| 1.4538 \| 0.1814 \| 1.4538 \| 1.2057 \| \| No log \| 6.1739 \| 284 \| 1.6059 \| 0.2694 \| 1.6059 \| 1.2673 \| \| No log \| 6.2174 \| 286 \| 1.5940 \| 0.2315 \| 1.5940 \| 1.2625 \| \| No log \| 6.2609 \| 288 \| 1.5500 \| 0.2270 \| 1.5500 \| 1.2450 \| \| No log \| 6.3043 \| 290 \| 1.4412 \| 0.1850 \| 1.4412 \| 1.2005 \| \| No log \| 6.3478 \| 292 \| 1.3873 \| 0.1769 \| 1.3873 \| 1.1778 \| \| No log \| 6.3913 \| 294 \| 1.2601 \| 0.0 \| 1.2601 \| 1.1226 \| \| No log \| 6.4348 \| 296 \| 1.2080 \| 0.0 \| 1.2080 \| 1.0991 \| \| No log \| 6.4783 \| 298 \| 1.2249 \| 0.0 \| 1.2249 \| 1.1067 \| \| No log \| 6.5217 \| 300 \| 1.2305 \| 0.0 \| 1.2305 \| 1.1093 \| \| No log \| 6.5652 \| 302 \| 1.2659 \| 0.0033 \| 1.2659 \| 1.1251 \| \| No log \| 6.6087 \| 304 \| 1.3705 \| 0.1255 \| 1.3705 \| 1.1707 \| \| No log \| 6.6522 \| 306 \| 1.5111 \| 0.1370 \| 1.5111 \| 1.2293 \| \| No log \| 6.6957 \| 308 \| 1.6192 \| 0.2342 \| 1.6192 \| 1.2725 \| \| No log \| 6.7391 \| 310 \| 1.5472 \| 0.2424 \| 1.5472 \| 1.2439 \| \| No log \| 6.7826 \| 312 \| 1.4260 \| 0.1113 \| 1.4260 \| 1.1942 \| \| No log \| 6.8261 \| 314 \| 1.3601 \| 0.0 \| 1.3601 \| 1.1662 \| \| No log \| 6.8696 \| 316 \| 1.4354 \| 0.0 \| 1.4354 \| 1.1981 \| \| No log \| 6.9130 \| 318 \| 1.4593 \| 0.0 \| 1.4593 \| 1.2080 \| \| No log \| 6.9565 \| 320 \| 1.5025 \| 0.0661 \| 1.5025 \| 1.2258 \| \| No log \| 7.0 \| 322 \| 1.5379 \| 0.1052 \| 1.5379 \| 1.2401 \| \| No log \| 7.0435 \| 324 \| 1.5436 \| 0.1744 \| 1.5436 \| 1.2424 \| \| No log \| 7.0870 \| 326 \| 1.5589 \| 0.0661 \| 1.5589 \| 1.2486 \| \| No log \| 7.1304 \| 328 \| 1.5395 \| 0.0883 \| 1.5395 \| 1.2408 \| \| No log \| 7.1739 \| 330 \| 1.5589 \| 0.1734 \| 1.5589 \| 1.2485 \| \| No log \| 7.2174 \| 332 \| 1.7114 \| 0.1808 \| 1.7114 \| 1.3082 \| \| No log \| 7.2609 \| 334 \| 1.8738 \| 0.1002 \| 1.8738 \| 1.3689 \| \| No log \| 7.3043 \| 336 \| 1.9387 \| 0.1002 \| 1.9387 \| 1.3924 \| \| No log \| 7.3478 \| 338 \| 1.8955 \| 0.1807 \| 1.8955 \| 1.3768 \| \| No log \| 7.3913 \| 340 \| 1.6579 \| 0.2117 \| 1.6579 \| 1.2876 \| \| No log \| 7.4348 \| 342 \| 1.4418 \| 0.1288 \| 1.4418 \| 1.2008 \| \| No log \| 7.4783 \| 344 \| 1.3060 \| 0.0931 \| 1.3060 \| 1.1428 \| \| No log \| 7.5217 \| 346 \| 1.3268 \| 0.0541 \| 1.3268 \| 1.1519 \| \| No log \| 7.5652 \| 348 \| 1.3956 \| 0.1744 \| 1.3956 \| 1.1813 \| \| No log \| 7.6087 \| 350 \| 1.5431 \| 0.2474 \| 1.5431 \| 1.2422 \| \| No log \| 7.6522 \| 352 \| 1.6069 \| 0.2611 \| 1.6069 \| 1.2676 \| \| No log \| 7.6957 \| 354 \| 1.6304 \| 0.2174 \| 1.6304 \| 1.2769 \| \| No log \| 7.7391 \| 356 \| 1.5382 \| 0.2123 \| 1.5382 \| 1.2402 \| \| No log \| 7.7826 \| 358 \| 1.3799 \| 0.1727 \| 1.3799 \| 1.1747 \| \| No log \| 7.8261 \| 360 \| 1.2919 \| 0.1316 \| 1.2919 \| 1.1366 \| \| No log \| 7.8696 \| 362 \| 1.3619 \| 0.1024 \| 1.3619 \| 1.1670 \| \| No log \| 7.9130 \| 364 \| 1.4500 \| 0.2239 \| 1.4500 \| 1.2041 \| \| No log \| 7.9565 \| 366 \| 1.4819 \| 0.2342 \| 1.4819 \| 1.2173 \| \| No log \| 8.0 \| 368 \| 1.3776 \| 0.1197 \| 1.3776 \| 1.1737 \| \| No log \| 8.0435 \| 370 \| 1.1953 \| 0.1019 \| 1.1953 \| 1.0933 \| \| No log \| 8.0870 \| 372 \| 1.1063 \| 0.0406 \| 1.1063 \| 1.0518 \| \| No log \| 8.1304 \| 374 \| 1.1168 \| 0.0406 \| 1.1168 \| 1.0568 \| \| No log \| 8.1739 \| 376 \| 1.2230 \| 0.1705 \| 1.2230 \| 1.1059 \| \| No log \| 8.2174 \| 378 \| 1.4200 \| 0.1370 \| 1.4200 \| 1.1916 \| \| No log \| 8.2609 \| 380 \| 1.4560 \| 0.0931 \| 1.4560 \| 1.2066 \| \| No log \| 8.3043 \| 382 \| 1.4686 \| 0.0931 \| 1.4686 \| 1.2119 \| \| No log \| 8.3478 \| 384 \| 1.4976 \| 0.1744 \| 1.4976 \| 1.2237 \| \| No log \| 8.3913 \| 386 \| 1.4849 \| 0.1744 \| 1.4849 \| 1.2186 \| \| No log \| 8.4348 \| 388 \| 1.5288 \| 0.2665 \| 1.5288 \| 1.2365 \| \| No log \| 8.4783 \| 390 \| 1.5550 \| 0.2752 \| 1.5550 \| 1.2470 \| \| No log \| 8.5217 \| 392 \| 1.4539 \| 0.2372 \| 1.4539 \| 1.2058 \| \| No log \| 8.5652 \| 394 \| 1.3677 \| 0.2065 \| 1.3677 \| 1.1695 \| \| No log \| 8.6087 \| 396 \| 1.3685 \| 0.1407 \| 1.3685 \| 1.1698 \| \| No log \| 8.6522 \| 398 \| 1.4109 \| 0.1744 \| 1.4109 \| 1.1878 \| \| No log \| 8.6957 \| 400 \| 1.5116 \| 0.2424 \| 1.5116 \| 1.2295 \| \| No log \| 8.7391 \| 402 \| 1.5118 \| 0.2474 \| 1.5118 \| 1.2295 \| \| No log \| 8.7826 \| 404 \| 1.5575 \| 0.2522 \| 1.5575 \| 1.2480 \| \| No log \| 8.8261 \| 406 \| 1.5527 \| 0.2126 \| 1.5527 \| 1.2461 \| \| No log \| 8.8696 \| 408 \| 1.4464 \| 0.1407 \| 1.4464 \| 1.2027 \| \| No log \| 8.9130 \| 410 \| 1.4063 \| 0.0 \| 1.4063 \| 1.1859 \| \| No log \| 8.9565 \| 412 \| 1.3768 \| 0.0 \| 1.3768 \| 1.1734 \| \| No log \| 9.0 \| 414 \| 1.3550 \| -0.0411 \| 1.3550 \| 1.1641 \| \| No log \| 9.0435 \| 416 \| 1.3643 \| 0.0 \| 1.3643 \| 1.1680 \| \| No log \| 9.0870 \| 418 \| 1.4810 \| 0.0401 \| 1.4810 \| 1.2170 \| \| No log \| 9.1304 \| 420 \| 1.6359 \| 0.2126 \| 1.6359 \| 1.2790 \| \| No log \| 9.1739 \| 422 \| 1.7412 \| 0.1892 \| 1.7412 \| 1.3196 \| \| No log \| 9.2174 \| 424 \| 1.8050 \| 0.1487 \| 1.8050 \| 1.3435 \| \| No log \| 9.2609 \| 426 \| 1.7355 \| 0.1955 \| 1.7355 \| 1.3174 \| \| No log \| 9.3043 \| 428 \| 1.5760 \| 0.2653 \| 1.5760 \| 1.2554 \| \| No log \| 9.3478 \| 430 \| 1.4390 \| 0.2015 \| 1.4390 \| 1.1996 \| \| No log \| 9.3913 \| 432 \| 1.3894 \| 0.1700 \| 1.3894 \| 1.1787 \| \| No log \| 9.4348 \| 434 \| 1.3953 \| 0.1744 \| 1.3953 \| 1.1812 \| \| No log \| 9.4783 \| 436 \| 1.4498 \| 0.1744 \| 1.4498 \| 1.2041 \| \| No log \| 9.5217 \| 438 \| 1.4705 \| 0.1744 \| 1.4705 \| 1.2126 \| \| No log \| 9.5652 \| 440 \| 1.3951 \| 0.1744 \| 1.3951 \| 1.1811 \| \| No log \| 9.6087 \| 442 \| 1.2963 \| 0.1052 \| 1.2963 \| 1.1385 \| \| No log \| 9.6522 \| 444 \| 1.2880 \| 0.1052 \| 1.2880 \| 1.1349 \| \| No log \| 9.6957 \| 446 \| 1.3230 \| 0.1407 \| 1.3230 \| 1.1502 \| \| No log \| 9.7391 \| 448 \| 1.3931 \| 0.1407 \| 1.3931 \| 1.1803 \| \| No log \| 9.7826 \| 450 \| 1.4239 \| 0.1052 \| 1.4239 \| 1.1933 \| \| No log \| 9.8261 \| 452 \| 1.4655 \| 0.1744 \| 1.4655 \| 1.2106 \| \| No log \| 9.8696 \| 454 \| 1.4982 \| 0.2372 \| 1.4982 \| 1.2240 \| \| No log \| 9.9130 \| 456 \| 1.5002 \| 0.2372 \| 1.5002 \| 1.2248 \| \| No log \| 9.9565 \| 458 \| 1.4132 \| 0.2372 \| 1.4132 \| 1.1888 \| \| No log \| 10.0 \| 460 \| 1.2400 \| 0.0 \| 1.2400 \| 1.1135 \| \| No log \| 10.0435 \| 462 \| 1.1297 \| 0.0155 \| 1.1297 \| 1.0629 \| \| No log \| 10.0870 \| 464 \| 1.1569 \| 0.0155 \| 1.1569 \| 1.0756 \| \| No log \| 10.1304 \| 466 \| 1.2550 \| 0.0 \| 1.2550 \| 1.1203 \| \| No log \| 10.1739 \| 468 \| 1.2879 \| 0.1113 \| 1.2879 \| 1.1348 \| \| No log \| 10.2174 \| 470 \| 1.3146 \| 0.0390 \| 1.3146 \| 1.1466 \| \| No log \| 10.2609 \| 472 \| 1.3950 \| 0.0781 \| 1.3950 \| 1.1811 \| \| No log \| 10.3043 \| 474 \| 1.4311 \| 0.0781 \| 1.4311 \| 1.1963 \| \| No log \| 10.3478 \| 476 \| 1.4025 \| 0.0781 \| 1.4025 \| 1.1843 \| \| No log \| 10.3913 \| 478 \| 1.3441 \| 0.0 \| 1.3441 \| 1.1593 \| \| No log \| 10.4348 \| 480 \| 1.3662 \| 0.0 \| 1.3662 \| 1.1689 \| \| No log \| 10.4783 \| 482 \| 1.4121 \| 0.1370 \| 1.4121 \| 1.1883 \| \| No log \| 10.5217 \| 484 \| 1.5203 \| 0.2015 \| 1.5203 \| 1.2330 \| \| No log \| 10.5652 \| 486 \| 1.7276 \| 0.1963 \| 1.7276 \| 1.3144 \| \| No log \| 10.6087 \| 488 \| 1.8236 \| 0.2056 \| 1.8236 \| 1.3504 \| \| No log \| 10.6522 \| 490 \| 1.7907 \| 0.2317 \| 1.7907 \| 1.3382 \| \| No log \| 10.6957 \| 492 \| 1.6974 \| 0.2391 \| 1.6974 \| 1.3028 \| \| No log \| 10.7391 \| 494 \| 1.5156 \| 0.1744 \| 1.5156 \| 1.2311 \| \| No log \| 10.7826 \| 496 \| 1.4285 \| 0.0390 \| 1.4285 \| 1.1952 \| \| No log \| 10.8261 \| 498 \| 1.3874 \| 0.0390 \| 1.3874 \| 1.1779 \| \| 0.2613 \| 10.8696 \| 500 \| 1.3506 \| 0.0 \| 1.3506 \| 1.1622 \| \| 0.2613 \| 10.9130 \| 502 \| 1.3637 \| 0.0661 \| 1.3637 \| 1.1678 \| \| 0.2613 \| 10.9565 \| 504 \| 1.4279 \| 0.1370 \| 1.4279 \| 1.1950 \| \| 0.2613 \| 11.0 \| 506 \| 1.5500 \| 0.2315 \| 1.5500 \| 1.2450 \| \| 0.2613 \| 11.0435 \| 508 \| 1.6369 \| 0.2465 \| 1.6369 \| 1.2794 \| \| 0.2613 \| 11.0870 \| 510 \| 1.7018 \| 0.2832 \| 1.7018 \| 1.3045 \| \| 0.2613 \| 11.1304 \| 512 \| 1.6085 \| 0.2665 \| 1.6085 \| 1.2683 \| \| 0.2613 \| 11.1739 \| 514 \| 1.4327 \| 0.1744 \| 1.4327 \| 1.1969 \| \| 0.2613 \| 11.2174 \| 516 \| 1.3146 \| 0.0401 \| 1.3146 \| 1.1466 \| \| 0.2613 \| 11.2609 \| 518 \| 1.2902 \| 0.0401 \| 1.2902 \| 1.1359 \| \| 0.2613 \| 11.3043 \| 520 \| 1.3147 \| 0.0401 \| 1.3147 \| 1.1466 \| \| 0.2613 \| 11.3478 \| 522 \| 1.3176 \| 0.1407 \| 1.3176 \| 1.1479 \| \| 0.2613 \| 11.3913 \| 524 \| 1.3261 \| 0.1744 \| 1.3261 \| 1.1516 \| \| 0.2613 \| 11.4348 \| 526 \| 1.2363 \| 0.2206 \| 1.2363 \| 1.1119 \| \| 0.2613 \| 11.4783 \| 528 \| 1.2522 \| 0.2313 \| 1.2522 \| 1.1190 \| \| 0.2613 \| 11.5217 \| 530 \| 1.4130 \| 0.2431 \| 1.4130 \| 1.1887 \| \| 0.2613 \| 11.5652 \| 532 \| 1.7543 \| 0.1911 \| 1.7543 \| 1.3245 \| \| 0.2613 \| 11.6087 \| 534 \| 1.8516 \| 0.2406 \| 1.8516 \| 1.3607 \| \| 0.2613 \| 11.6522 \| 536 \| 1.7143 \| 0.1892 \| 1.7143 \| 1.3093 \| \| 0.2613 \| 11.6957 \| 538 \| 1.5454 \| 0.1288 \| 1.5454 \| 1.2431 \| \| 0.2613 \| 11.7391 \| 540 \| 1.4371 \| 0.0781 \| 1.4371 \| 1.1988 \| \| 0.2613 \| 11.7826 \| 542 \| 1.3778 \| 0.1370 \| 1.3778 \| 1.1738 \| \| 0.2613 \| 11.8261 \| 544 \| 1.3337 \| 0.1370 \| 1.3337 \| 1.1549 \| \| 0.2613 \| 11.8696 \| 546 \| 1.3540 \| 0.1700 \| 1.3540 \| 1.1636 \| \| 0.2613 \| 11.9130 \| 548 \| 1.4217 \| 0.2126 \| 1.4217 \| 1.1924 \| \| 0.2613 \| 11.9565 \| 550 \| 1.3881 \| 0.2065 \| 1.3881 \| 1.1782 \| \| 0.2613 \| 12.0 \| 552 \| 1.2795 \| 0.0781 \| 1.2795 \| 1.1312 \| \| 0.2613 \| 12.0435 \| 554 \| 1.1736 \| 0.0155 \| 1.1736 \| 1.0834 \| \| 0.2613 \| 12.0870 \| 556 \| 1.1141 \| 0.0587 \| 1.1141 \| 1.0555 \| \| 0.2613 \| 12.1304 \| 558 \| 1.0583 \| 0.0618 \| 1.0583 \| 1.0288 \| \| 0.2613 \| 12.1739 \| 560 \| 1.0693 \| 0.1407 \| 1.0693 \| 1.0341 \| \| 0.2613 \| 12.2174 \| 562 \| 1.1314 \| 0.0961 \| 1.1314 \| 1.0637 \| \| 0.2613 \| 12.2609 \| 564 \| 1.2606 \| 0.1769 \| 1.2606 \| 1.1228 \| \| 0.2613 \| 12.3043 \| 566 \| 1.3739 \| 0.2126 \| 1.3739 \| 1.1721 \| \| 0.2613 \| 12.3478 \| 568 \| 1.4466 \| 0.2126 \| 1.4466 \| 1.2028 \| \| 0.2613 \| 12.3913 \| 570 \| 1.4543 \| 0.2126 \| 1.4543 \| 1.2060 \| \| 0.2613 \| 12.4348 \| 572 \| 1.5068 \| 0.2424 \| 1.5068 \| 1.2275 \| \| 0.2613 \| 12.4783 \| 574 \| 1.6092 \| 0.2832 \| 1.6092 \| 1.2685 \| \| 0.2613 \| 12.5217 \| 576 \| 1.7369 \| 0.2940 \| 1.7369 \| 1.3179 \| \| 0.2613 \| 12.5652 \| 578 \| 1.7533 \| 0.2940 \| 1.7533 \| 1.3241 \| \| 0.2613 \| 12.6087 \| 580 \| 1.6560 \| 0.2474 \| 1.6560 \| 1.2868 \| \| 0.2613 \| 12.6522 \| 582 \| 1.5699 \| 0.1228 \| 1.5699 \| 1.2530 \| \| 0.2613 \| 12.6957 \| 584 \| 1.5091 \| 0.0781 \| 1.5091 \| 1.2285 \| \| 0.2613 \| 12.7391 \| 586 \| 1.4348 \| 0.0401 \| 1.4348 \| 1.1978 \| \| 0.2613 \| 12.7826 \| 588 \| 1.3822 \| 0.0401 \| 1.3822 \| 1.1757 \| \| 0.2613 \| 12.8261 \| 590 \| 1.3843 \| 0.0401 \| 1.3843 \| 1.1766 \| \| 0.2613 \| 12.8696 \| 592 \| 1.4302 \| 0.1407 \| 1.4302 \| 1.1959 \| \| 0.2613 \| 12.9130 \| 594 \| 1.4860 \| 0.2424 \| 1.4860 \| 1.2190 \| \| 0.2613 \| 12.9565 \| 596 \| 1.5159 \| 0.2474 \| 1.5159 \| 1.2312 \| \| 0.2613 \| 13.0 \| 598 \| 1.5296 \| 0.2793 \| 1.5296 \| 1.2368 \| \| 0.2613 \| 13.0435 \| 600 \| 1.6018 \| 0.2832 \| 1.6018 \| 1.2656 \| \| 0.2613 \| 13.0870 \| 602 \| 1.5336 \| 0.2832 \| 1.5336 \| 1.2384 \| \| 0.2613 \| 13.1304 \| 604 \| 1.4550 \| 0.2474 \| 1.4550 \| 1.2062 \| \| 0.2613 \| 13.1739 \| 606 \| 1.4058 \| 0.2424 \| 1.4058 \| 1.1856 \| \| 0.2613 \| 13.2174 \| 608 \| 1.3353 \| 0.2075 \| 1.3353 \| 1.1556 \| \| 0.2613 \| 13.2609 \| 610 \| 1.3624 \| 0.2075 \| 1.3624 \| 1.1672 \| \| 0.2613 \| 13.3043 \| 612 \| 1.3326 \| 0.1769 \| 1.3326 \| 1.1544 \| \| 0.2613 \| 13.3478 \| 614 \| 1.3241 \| 0.1769 \| 1.3241 \| 1.1507 \| \| 0.2613 \| 13.3913 \| 616 \| 1.3316 \| 0.1769 \| 1.3316 \| 1.1539 \| \| 0.2613 \| 13.4348 \| 618 \| 1.3209 \| 0.1769 \| 1.3209 \| 1.1493 \| \| 0.2613 \| 13.4783 \| 620 \| 1.3151 \| 0.1769 \| 1.3151 \| 1.1468 \| \| 0.2613 \| 13.5217 \| 622 \| 1.3106 \| 0.1769 \| 1.3106 \| 1.1448 \| \| 0.2613 \| 13.5652 \| 624 \| 1.3274 \| 0.2424 \| 1.3274 \| 1.1521 \| \| 0.2613 \| 13.6087 \| 626 \| 1.4479 \| 0.2752 \| 1.4479 \| 1.2033 \| \| 0.2613 \| 13.6522 \| 628 \| 1.5377 \| 0.2568 \| 1.5377 \| 1.2400 \| \| 0.2613 \| 13.6957 \| 630 \| 1.5216 \| 0.2568 \| 1.5216 \| 1.2336 \| \| 0.2613 \| 13.7391 \| 632 \| 1.5517 \| 0.2568 \| 1.5517 \| 1.2457 \| \| 0.2613 \| 13.7826 \| 634 \| 1.6063 \| 0.2568 \| 1.6063 \| 1.2674 \| \| 0.2613 \| 13.8261 \| 636 \| 1.5487 \| 0.2752 \| 1.5487 \| 1.2445 \| \| 0.2613 \| 13.8696 \| 638 \| 1.4303 \| 0.2065 \| 1.4303 \| 1.1960 \| \| 0.2613 \| 13.9130 \| 640 \| 1.3684 \| 0.2065 \| 1.3684 \| 1.1698 \| \| 0.2613 \| 13.9565 \| 642 \| 1.3211 \| 0.0401 \| 1.3211 \| 1.1494 \| \| 0.2613 \| 14.0 \| 644 \| 1.3409 \| 0.0401 \| 1.3409 \| 1.1580 \| \| 0.2613 \| 14.0435 \| 646 \| 1.4071 \| 0.0781 \| 1.4071 \| 1.1862 \| \| 0.2613 \| 14.0870 \| 648 \| 1.4992 \| 0.2065 \| 1.4992 \| 1.2244 \| \| 0.2613 \| 14.1304 \| 650 \| 1.6000 \| 0.2709 \| 1.6000 \| 1.2649 \| \| 0.2613 \| 14.1739 \| 652 \| 1.6450 \| 0.2752 \| 1.6450 \| 1.2826 \| \| 0.2613 \| 14.2174 \| 654 \| 1.6421 \| 0.2752 \| 1.6421 \| 1.2814 \| \| 0.2613 \| 14.2609 \| 656 \| 1.6102 \| 0.2417 \| 1.6102 \| 1.2689 \| \| 0.2613 \| 14.3043 \| 658 \| 1.5413 \| 0.2372 \| 1.5413 \| 1.2415 \| \| 0.2613 \| 14.3478 \| 660 \| 1.5115 \| 0.2372 \| 1.5115 \| 1.2294 \| \| 0.2613 \| 14.3913 \| 662 \| 1.5279 \| 0.2372 \| 1.5279 \| 1.2361 \| \| 0.2613 \| 14.4348 \| 664 \| 1.5128 \| 0.2372 \| 1.5128 \| 1.2300 \| \| 0.2613 \| 14.4783 \| 666 \| 1.5337 \| 0.2424 \| 1.5337 \| 1.2384 \| \| 0.2613 \| 14.5217 \| 668 \| 1.5367 \| 0.2752 \| 1.5367 \| 1.2397 \| \| 0.2613 \| 14.5652 \| 670 \| 1.5846 \| 0.2752 \| 1.5846 \| 1.2588 \| \| 0.2613 \| 14.6087 \| 672 \| 1.6122 \| 0.2832 \| 1.6122 \| 1.2697 \| \| 0.2613 \| 14.6522 \| 674 \| 1.5983 \| 0.2752 \| 1.5983 \| 1.2642 \| \| 0.2613 \| 14.6957 \| 676 \| 1.5849 \| 0.2752 \| 1.5849 \| 1.2589 \| \| 0.2613 \| 14.7391 \| 678 \| 1.5418 \| 0.2752 \| 1.5418 \| 1.2417 \| \| 0.2613 \| 14.7826 \| 680 \| 1.4476 \| 0.1744 \| 1.4476 \| 1.2032 \| \| 0.2613 \| 14.8261 \| 682 \| 1.3866 \| 0.1142 \| 1.3866 \| 1.1776 \| \| 0.2613 \| 14.8696 \| 684 \| 1.3755 \| 0.0401 \| 1.3755 \| 1.1728 \| \| 0.2613 \| 14.9130 \| 686 \| 1.3934 \| 0.0781 \| 1.3934 \| 1.1804 \| \| 0.2613 \| 14.9565 \| 688 \| 1.4456 \| 0.1142 \| 1.4456 \| 1.2023 \| \| 0.2613 \| 15.0 \| 690 \| 1.5034 \| 0.1769 \| 1.5034 \| 1.2261 \| \| 0.2613 \| 15.0435 \| 692 \| 1.5961 \| 0.2752 \| 1.5961 \| 1.2634 \| \| 0.2613 \| 15.0870 \| 694 \| 1.7191 \| 0.2832 \| 1.7191 \| 1.3111 \| \| 0.2613 \| 15.1304 \| 696 \| 1.8074 \| 0.2406 \| 1.8074 \| 1.3444 \| \| 0.2613 \| 15.1739 \| 698 \| 1.7995 \| 0.2770 \| 1.7995 \| 1.3414 \| \| 0.2613 \| 15.2174 \| 700 \| 1.6816 \| 0.2752 \| 1.6816 \| 1.2968 \| \| 0.2613 \| 15.2609 \| 702 \| 1.5217 \| 0.1142 \| 1.5217 \| 1.2336 \| \| 0.2613 \| 15.3043 \| 704 \| 1.3637 \| 0.0401 \| 1.3637 \| 1.1678 \| \| 0.2613 \| 15.3478 \| 706 \| 1.2650 \| 0.0 \| 1.2650 \| 1.1247 \| \| 0.2613 \| 15.3913 \| 708 \| 1.2534 \| 0.0 \| 1.2534 \| 1.1196 \| \| 0.2613 \| 15.4348 \| 710 \| 1.2924 \| 0.0 \| 1.2924 \| 1.1368 \| \| 0.2613 \| 15.4783 \| 712 \| 1.3811 \| 0.0401 \| 1.3811 \| 1.1752 \| \| 0.2613 \| 15.5217 \| 714 \| 1.5260 \| 0.0781 \| 1.5260 \| 1.2353 \| \| 0.2613 \| 15.5652 \| 716 \| 1.6227 \| 0.2065 \| 1.6227 \| 1.2738 \| \| 0.2613 \| 15.6087 \| 718 \| 1.6985 \| 0.2709 \| 1.6985 \| 1.3033 \| \| 0.2613 \| 15.6522 \| 720 \| 1.7279 \| 0.2709 \| 1.7279 \| 1.3145 \| \| 0.2613 \| 15.6957 \| 722 \| 1.6787 \| 0.0781 \| 1.6787 \| 1.2956 \| \| 0.2613 \| 15.7391 \| 724 \| 1.5888 \| 0.0401 \| 1.5888 \| 1.2605 \| \| 0.2613 \| 15.7826 \| 726 \| 1.5083 \| 0.0 \| 1.5083 \| 1.2281 \| \| 0.2613 \| 15.8261 \| 728 \| 1.4668 \| 0.0 \| 1.4668 \| 1.2111 \| \| 0.2613 \| 15.8696 \| 730 \| 1.4395 \| 0.0 \| 1.4395 \| 1.1998 \| \| 0.2613 \| 15.9130 \| 732 \| 1.4284 \| 0.0401 \| 1.4284 \| 1.1952 \| \| 0.2613 \| 15.9565 \| 734 \| 1.4698 \| 0.0401 \| 1.4698 \| 1.2123 \| \| 0.2613 \| 16.0 \| 736 \| 1.4808 \| 0.0781 \| 1.4808 \| 1.2169 \| \| 0.2613 \| 16.0435 \| 738 \| 1.4348 \| 0.0401 \| 1.4348 \| 1.1978 \| \| 0.2613 \| 16.0870 \| 740 \| 1.3826 \| 0.0401 \| 1.3826 \| 1.1758 \| \| 0.2613 \| 16.1304 \| 742 \| 1.3705 \| 0.0401 \| 1.3705 \| 1.1707 \| \| 0.2613 \| 16.1739 \| 744 \| 1.4221 \| 0.0401 \| 1.4221 \| 1.1925 \| \| 0.2613 \| 16.2174 \| 746 \| 1.5149 \| 0.0781 \| 1.5149 \| 1.2308 \| \| 0.2613 \| 16.2609 \| 748 \| 1.6344 \| 0.1142 \| 1.6344 \| 1.2785 \| \| 0.2613 \| 16.3043 \| 750 \| 1.6958 \| 0.1142 \| 1.6958 \| 1.3022 \| \| 0.2613 \| 16.3478 \| 752 \| 1.7174 \| 0.1744 \| 1.7174 \| 1.3105 \| \| 0.2613 \| 16.3913 \| 754 \| 1.6890 \| 0.1744 \| 1.6890 \| 1.2996 \| \| 0.2613 \| 16.4348 \| 756 \| 1.6066 \| 0.1142 \| 1.6066 \| 1.2675 \| \| 0.2613 \| 16.4783 \| 758 \| 1.5513 \| 0.0390 \| 1.5513 \| 1.2455 \| \| 0.2613 \| 16.5217 \| 760 \| 1.5064 \| 0.0390 \| 1.5064 \| 1.2273 \| \| 0.2613 \| 16.5652 \| 762 \| 1.4922 \| 0.0390 \| 1.4922 \| 1.2215 \| \| 0.2613 \| 16.6087 \| 764 \| 1.4445 \| 0.0390 \| 1.4445 \| 1.2019 \| \| 0.2613 \| 16.6522 \| 766 \| 1.4327 \| 0.1407 \| 1.4327 \| 1.1970 \| \| 0.2613 \| 16.6957 \| 768 \| 1.4487 \| 0.2065 \| 1.4487 \| 1.2036 \| \| 0.2613 \| 16.7391 \| 770 \| 1.3882 \| 0.0781 \| 1.3882 \| 1.1782 \| \| 0.2613 \| 16.7826 \| 772 \| 1.2968 \| 0.0781 \| 1.2968 \| 1.1388 \| \| 0.2613 \| 16.8261 \| 774 \| 1.2913 \| 0.0781 \| 1.2913 \| 1.1363 \| \| 0.2613 \| 16.8696 \| 776 \| 1.3576 \| 0.2372 \| 1.3576 \| 1.1652 \| \| 0.2613 \| 16.9130 \| 778 \| 1.4646 \| 0.2424 \| 1.4646 \| 1.2102 \| \| 0.2613 \| 16.9565 \| 780 \| 1.4935 \| 0.2522 \| 1.4935 \| 1.2221 \| \| 0.2613 \| 17.0 \| 782 \| 1.4955 \| 0.2709 \| 1.4955 \| 1.2229 \| \| 0.2613 \| 17.0435 \| 784 \| 1.4958 \| 0.2075 \| 1.4958 \| 1.2230 \| \| 0.2613 \| 17.0870 \| 786 \| 1.4286 \| 0.1113 \| 1.4286 \| 1.1952 \| \| 0.2613 \| 17.1304 \| 788 \| 1.3892 \| 0.0155 \| 1.3892 \| 1.1786 \| \| 0.2613 \| 17.1739 \| 790 \| 1.3827 \| 0.0155 \| 1.3827 \| 1.1759 \| \| 0.2613 \| 17.2174 \| 792 \| 1.3759 \| 0.0155 \| 1.3759 \| 1.1730 \| \| 0.2613 \| 17.2609 \| 794 \| 1.4105 \| 0.0155 \| 1.4105 \| 1.1876 \| \| 0.2613 \| 17.3043 \| 796 \| 1.4572 \| 0.1113 \| 1.4572 \| 1.2071 \| \| 0.2613 \| 17.3478 \| 798 \| 1.4896 \| 0.2424 \| 1.4896 \| 1.2205 \| \| 0.2613 \| 17.3913 \| 800 \| 1.4728 \| 0.2424 \| 1.4728 \| 1.2136 \| \| 0.2613 \| 17.4348 \| 802 \| 1.4114 \| 0.2126 \| 1.4114 \| 1.1880 \| \| 0.2613 \| 17.4783 \| 804 \| 1.3548 \| 0.0781 \| 1.3548 \| 1.1640 \| \| 0.2613 \| 17.5217 \| 806 \| 1.3085 \| 0.0401 \| 1.3085 \| 1.1439 \| \| 0.2613 \| 17.5652 \| 808 \| 1.2499 \| 0.0 \| 1.2499 \| 1.1180 \| \| 0.2613 \| 17.6087 \| 810 \| 1.2468 \| 0.0 \| 1.2468 \| 1.1166 \| \| 0.2613 \| 17.6522 \| 812 \| 1.3334 \| 0.0401 \| 1.3334 \| 1.1547 \| \| 0.2613 \| 17.6957 \| 814 \| 1.4046 \| 0.2424 \| 1.4046 \| 1.1852 \| \| 0.2613 \| 17.7391 \| 816 \| 1.4401 \| 0.2424 \| 1.4401 \| 1.2000 \| \| 0.2613 \| 17.7826 \| 818 \| 1.4071 \| 0.2075 \| 1.4071 \| 1.1862 \| \| 0.2613 \| 17.8261 \| 820 \| 1.4065 \| 0.2367 \| 1.4065 \| 1.1860 \| \| 0.2613 \| 17.8696 \| 822 \| 1.3824 \| 0.2367 \| 1.3824 \| 1.1758 \| \| 0.2613 \| 17.9130 \| 824 \| 1.3945 \| 0.2367 \| 1.3945 \| 1.1809 \| \| 0.2613 \| 17.9565 \| 826 \| 1.4377 \| 0.2647 \| 1.4377 \| 1.1990 \| \| 0.2613 \| 18.0 \| 828 \| 1.4406 \| 0.2015 \| 1.4406 \| 1.2002 \| \| 0.2613 \| 18.0435 \| 830 \| 1.4659 \| 0.1370 \| 1.4659 \| 1.2108 \| \| 0.2613 \| 18.0870 \| 832 \| 1.4324 \| 0.1370 \| 1.4324 \| 1.1968 \| \| 0.2613 \| 18.1304 \| 834 \| 1.3917 \| 0.0390 \| 1.3917 \| 1.1797 \| \| 0.2613 \| 18.1739 \| 836 \| 1.3690 \| 0.1024 \| 1.3690 \| 1.1700 \| \| 0.2613 \| 18.2174 \| 838 \| 1.3763 \| 0.1370 \| 1.3763 \| 1.1732 \| \| 0.2613 \| 18.2609 \| 840 \| 1.4114 \| 0.1449 \| 1.4114 \| 1.1880 \| \| 0.2613 \| 18.3043 \| 842 \| 1.4059 \| 0.1769 \| 1.4059 \| 1.1857 \| \| 0.2613 \| 18.3478 \| 844 \| 1.4071 \| 0.2038 \| 1.4071 \| 1.1862 \| \| 0.2613 \| 18.3913 \| 846 \| 1.4425 \| 0.1663 \| 1.4425 \| 1.2010 \| \| 0.2613 \| 18.4348 \| 848 \| 1.5222 \| 0.2291 \| 1.5222 \| 1.2338 \| \| 0.2613 \| 18.4783 \| 850 \| 1.4608 \| 0.1769 \| 1.4608 \| 1.2086 \| \| 0.2613 \| 18.5217 \| 852 \| 1.3935 \| 0.0390 \| 1.3935 \| 1.1805 \| \| 0.2613 \| 18.5652 \| 854 \| 1.3725 \| 0.0 \| 1.3725 \| 1.1715 \| \| 0.2613 \| 18.6087 \| 856 \| 1.3666 \| 0.0390 \| 1.3666 \| 1.1690 \| \| 0.2613 \| 18.6522 \| 858 \| 1.3549 \| 0.0760 \| 1.3549 \| 1.1640 \| \| 0.2613 \| 18.6957 \| 860 \| 1.3186 \| 0.0760 \| 1.3186 \| 1.1483 \| \| 0.2613 \| 18.7391 \| 862 \| 1.2866 \| 0.0760 \| 1.2866 \| 1.1343 \| \| 0.2613 \| 18.7826 \| 864 \| 1.2285 \| 0.0390 \| 1.2285 \| 1.1084 \| \| 0.2613 \| 18.8261 \| 866 \| 1.2261 \| 0.0 \| 1.2261 \| 1.1073 \| \| 0.2613 \| 18.8696 \| 868 \| 1.2744 \| 0.0 \| 1.2744 \| 1.1289 \| \| 0.2613 \| 18.9130 \| 870 \| 1.3796 \| 0.2065 \| 1.3796 \| 1.1746 \| \| 0.2613 \| 18.9565 \| 872 \| 1.4450 \| 0.2065 \| 1.4450 \| 1.2021 \| \| 0.2613 \| 19.0 \| 874 \| 1.4751 \| 0.2126 \| 1.4751 \| 1.2146 \| \| 0.2613 \| 19.0435 \| 876 \| 1.5427 \| 0.2424 \| 1.5427 \| 1.2420 \| \| 0.2613 \| 19.0870 \| 878 \| 1.5107 \| 0.2424 \| 1.5107 \| 1.2291 \| \| 0.2613 \| 19.1304 \| 880 \| 1.4398 \| 0.2065 \| 1.4398 \| 1.1999 \| \| 0.2613 \| 19.1739 \| 882 \| 1.3891 \| 0.2065 \| 1.3891 \| 1.1786 \| \| 0.2613 \| 19.2174 \| 884 \| 1.3882 \| 0.2065 \| 1.3882 \| 1.1782 \| \| 0.2613 \| 19.2609 \| 886 \| 1.3677 \| 0.2372 \| 1.3677 \| 1.1695 \| \| 0.2613 \| 19.3043 \| 888 \| 1.3458 \| 0.2372 \| 1.3458 \| 1.1601 \| \| 0.2613 \| 19.3478 \| 890 \| 1.3543 \| 0.2065 \| 1.3543 \| 1.1638 \| \| 0.2613 \| 19.3913 \| 892 \| 1.4147 \| 0.2065 \| 1.4147 \| 1.1894 \| \| 0.2613 \| 19.4348 \| 894 \| 1.4954 \| 0.2065 \| 1.4954 \| 1.2229 \| \| 0.2613 \| 19.4783 \| 896 \| 1.5624 \| 0.2372 \| 1.5624 \| 1.2500 \| \| 0.2613 \| 19.5217 \| 898 \| 1.5194 \| 0.2065 \| 1.5194 \| 1.2326 \| \| 0.2613 \| 19.5652 \| 900 \| 1.4135 \| 0.1142 \| 1.4135 \| 1.1889 \| \| 0.2613 \| 19.6087 \| 902 \| 1.3467 \| 0.0760 \| 1.3467 \| 1.1605 \| \| 0.2613 \| 19.6522 \| 904 \| 1.3329 \| 0.0760 \| 1.3329 \| 1.1545 \| \| 0.2613 \| 19.6957 \| 906 \| 1.3788 \| 0.0760 \| 1.3788 \| 1.1742 \| \| 0.2613 \| 19.7391 \| 908 \| 1.3941 \| 0.1142 \| 1.3941 \| 1.1807 \| \| 0.2613 \| 19.7826 \| 910 \| 1.3446 \| 0.0760 \| 1.3446 \| 1.1596 \| \| 0.2613 \| 19.8261 \| 912 \| 1.3464 \| 0.0781 \| 1.3464 \| 1.1603 \| \| 0.2613 \| 19.8696 \| 914 \| 1.3786 \| 0.0781 \| 1.3786 \| 1.1741 \| \| 0.2613 \| 19.9130 \| 916 \| 1.3890 \| 0.0781 \| 1.3890 \| 1.1786 \| \| 0.2613 \| 19.9565 \| 918 \| 1.4255 \| 0.1142 \| 1.4255 \| 1.1940 \| \| 0.2613 \| 20.0 \| 920 \| 1.4037 \| 0.1142 \| 1.4037 \| 1.1848 \| \| 0.2613 \| 20.0435 \| 922 \| 1.3277 \| 0.0760 \| 1.3277 \| 1.1523 \| \| 0.2613 \| 20.0870 \| 924 \| 1.3014 \| 0.1700 \| 1.3014 \| 1.1408 \| \| 0.2613 \| 20.1304 \| 926 \| 1.3892 \| 0.2075 \| 1.3892 \| 1.1786 \| \| 0.2613 \| 20.1739 \| 928 \| 1.5929 \| 0.2317 \| 1.5929 \| 1.2621 \| \| 0.2613 \| 20.2174 \| 930 \| 1.6970 \| 0.2317 \| 1.6970 \| 1.3027 \| \| 0.2613 \| 20.2609 \| 932 \| 1.7191 \| 0.2363 \| 1.7191 \| 1.3112 \| \| 0.2613 \| 20.3043 \| 934 \| 1.6680 \| 0.2568 \| 1.6680 \| 1.2915 \| \| 0.2613 \| 20.3478 \| 936 \| 1.6164 \| 0.2239 \| 1.6164 \| 1.2714 \| \| 0.2613 \| 20.3913 \| 938 \| 1.5876 \| 0.2372 \| 1.5876 \| 1.2600 \| \| 0.2613 \| 20.4348 \| 940 \| 1.5170 \| 0.1744 \| 1.5170 \| 1.2317 \| \| 0.2613 \| 20.4783 \| 942 \| 1.4762 \| 0.0781 \| 1.4762 \| 1.2150 \| \| 0.2613 \| 20.5217 \| 944 \| 1.4673 \| 0.0781 \| 1.4673 \| 1.2113 \| \| 0.2613 \| 20.5652 \| 946 \| 1.4335 \| 0.0781 \| 1.4335 \| 1.1973 \| \| 0.2613 \| 20.6087 \| 948 \| 1.4069 \| 0.0390 \| 1.4069 \| 1.1861 \| \| 0.2613 \| 20.6522 \| 950 \| 1.3963 \| 0.0390 \| 1.3963 \| 1.1816 \| \| 0.2613 \| 20.6957 \| 952 \| 1.3900 \| 0.0781 \| 1.3900 \| 1.1790 \| \| 0.2613 \| 20.7391 \| 954 \| 1.4040 \| 0.0781 \| 1.4040 \| 1.1849 \| \| 0.2613 \| 20.7826 \| 956 \| 1.4156 \| 0.1142 \| 1.4156 \| 1.1898 \| \| 0.2613 \| 20.8261 \| 958 \| 1.4111 \| 0.2126 \| 1.4111 \| 1.1879 \| \| 0.2613 \| 20.8696 \| 960 \| 1.4425 \| 0.2474 \| 1.4425 \| 1.2011 \| \| 0.2613 \| 20.9130 \| 962 \| 1.4642 \| 0.2474 \| 1.4642 \| 1.2100 \| \| 0.2613 \| 20.9565 \| 964 \| 1.4979 \| 0.2474 \| 1.4979 \| 1.2239 \| \| 0.2613 \| 21.0 \| 966 \| 1.4797 \| 0.2474 \| 1.4797 \| 1.2164 \| \| 0.2613 \| 21.0435 \| 968 \| 1.4519 \| 0.0878 \| 1.4519 \| 1.2049 \| \| 0.2613 \| 21.0870 \| 970 \| 1.4357 \| 0.0401 \| 1.4357 \| 1.1982 \| \| 0.2613 \| 21.1304 \| 972 \| 1.4458 \| 0.0510 \| 1.4458 \| 1.2024 \| \| 0.2613 \| 21.1739 \| 974 \| 1.4380 \| 0.1228 \| 1.4380 \| 1.1992 \| \| 0.2613 \| 21.2174 \| 976 \| 1.4340 \| 0.0781 \| 1.4340 \| 1.1975 \| \| 0.2613 \| 21.2609 \| 978 \| 1.4431 \| 0.0781 \| 1.4431 \| 1.2013 \| \| 0.2613 \| 21.3043 \| 980 \| 1.4430 \| 0.0781 \| 1.4430 \| 1.2012 \| \| 0.2613 \| 21.3478 \| 982 \| 1.4556 \| 0.0781 \| 1.4556 \| 1.2065 \| \| 0.2613 \| 21.3913 \| 984 \| 1.4817 \| 0.1142 \| 1.4817 \| 1.2172 \| \| 0.2613 \| 21.4348 \| 986 \| 1.4795 \| 0.1142 \| 1.4795 \| 1.2164 \| \| 0.2613 \| 21.4783 \| 988 \| 1.4291 \| 0.0390 \| 1.4291 \| 1.1955 \| \| 0.2613 \| 21.5217 \| 990 \| 1.3709 \| 0.0 \| 1.3709 \| 1.1709 \| \| 0.2613 \| 21.5652 \| 992 \| 1.3485 \| 0.0 \| 1.3485 \| 1.1613 \| \| 0.2613 \| 21.6087 \| 994 \| 1.3507 \| 0.0 \| 1.3507 \| 1.1622 \| \| 0.2613 \| 21.6522 \| 996 \| 1.3278 \| 0.0 \| 1.3278 \| 1.1523 \| \| 0.2613 \| 21.6957 \| 998 \| 1.3225 \| 0.0 \| 1.3225 \| 1.1500 \| ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1
thaffggg/94add4fb-1b6b-498e-bdda-09689212b56f	thaffggg	2025-01-20T23:40:46Z	6	0	peft	[ "peft", "safetensors", "phi3", "axolotl", "generated_from_trainer", "custom_code", "base_model:microsoft/Phi-3-mini-4k-instruct", "base_model:adapter:microsoft/Phi-3-mini-4k-instruct", "license:mit", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:10:03Z	--- library_name: peft license: mit base_model: microsoft/Phi-3-mini-4k-instruct tags: - axolotl - generated_from_trainer model-index: - name: 94add4fb-1b6b-498e-bdda-09689212b56f results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: microsoft/Phi-3-mini-4k-instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 2ae5e8a9f5305db8_train_data.json ds_type: json format: custom path: /workspace/input_data/2ae5e8a9f5305db8_train_data.json type: field_instruction: cleaned_description field_output: title format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: thaffggg/94add4fb-1b6b-498e-bdda-09689212b56f hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/2ae5e8a9f5305db8_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 5dae0642-01e7-4e34-8316-2fb97377e93c wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 5dae0642-01e7-4e34-8316-2fb97377e93c warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 94add4fb-1b6b-498e-bdda-09689212b56f This model is a fine-tuned version of [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.5853 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.4551 \| 0.0023 \| 200 \| 0.5853 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
LHRuig/realfoto	LHRuig	2025-01-20T23:40:19Z	10	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:39:14Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # realfoto <Gallery /> ## Model description realfoto lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/realfoto/tree/main) them in the Files & versions tab.
thabel/whisper-medium-yo	thabel	2025-01-20T23:40:06Z	98	0	transformers	[ "transformers", "tensorboard", "safetensors", "whisper", "automatic-speech-recognition", "generated_from_trainer", "dv", "dataset:mozilla-foundation/common_voice_13_0", "base_model:openai/whisper-medium", "base_model:finetune:openai/whisper-medium", "license:apache-2.0", "model-index", "endpoints_compatible", "region:us" ]	automatic-speech-recognition	2024-12-20T11:34:37Z	--- library_name: transformers language: - dv license: apache-2.0 base_model: openai/whisper-medium tags: - generated_from_trainer datasets: - mozilla-foundation/common_voice_13_0 metrics: - wer model-index: - name: Whisper Small Dv - Sanchit Gandhi results: - task: name: Automatic Speech Recognition type: automatic-speech-recognition dataset: name: Common Voice 13 type: mozilla-foundation/common_voice_13_0 config: yo split: test args: yo metrics: - name: Wer type: wer value: 47.33077228772023 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # Whisper Small Dv - Sanchit Gandhi This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Common Voice 13 dataset. It achieves the following results on the evaluation set: - Loss: 0.8186 - Wer Ortho: 69.8365 - Wer: 47.3308 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: constant_with_warmup - lr_scheduler_warmup_steps: 50 - training_steps: 500 - mixed_precision_training: Native AMP ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Wer Ortho \| Wer \| \|:-------------:\|:------:\|:----:\|:---------------:\|:---------:\|:-------:\| \| 0.061 \| 5.8824 \| 500 \| 0.8186 \| 69.8365 \| 47.3308 \| ### Framework versions - Transformers 4.47.1 - Pytorch 2.5.1+cu121 - Datasets 3.2.0 - Tokenizers 0.21.0
mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF	mradermacher	2025-01-20T23:39:50Z	504	1	transformers	[ "transformers", "gguf", "text-generation-inference", "unsloth", "gemma2", "trl", "sft", "en", "id", "dataset:gmonsoon/CoT-id", "base_model:gmonsoon/Eunoia-Gemma-9B-o1-Indo", "base_model:quantized:gmonsoon/Eunoia-Gemma-9B-o1-Indo", "license:gemma", "endpoints_compatible", "region:us", "conversational" ]	null	2025-01-14T14:09:42Z	--- base_model: gmonsoon/Eunoia-Gemma-9B-o1-Indo datasets: - gmonsoon/CoT-id language: - en - id library_name: transformers license: gemma quantized_by: mradermacher tags: - text-generation-inference - transformers - unsloth - gemma2 - trl - sft --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/gmonsoon/Eunoia-Gemma-9B-o1-Indo <!-- provided-files --> weighted/imatrix quants are available at https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q2_K.gguf) \| Q2_K \| 3.9 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q3_K_S.gguf) \| Q3_K_S \| 4.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q3_K_M.gguf) \| Q3_K_M \| 4.9 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q3_K_L.gguf) \| Q3_K_L \| 5.2 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.IQ4_XS.gguf) \| IQ4_XS \| 5.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q4_K_S.gguf) \| Q4_K_S \| 5.6 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q4_K_M.gguf) \| Q4_K_M \| 5.9 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q5_K_S.gguf) \| Q5_K_S \| 6.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q5_K_M.gguf) \| Q5_K_M \| 6.7 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q6_K.gguf) \| Q6_K \| 7.7 \| very good quality \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q8_0.gguf) \| Q8_0 \| 9.9 \| fast, best quality \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.f16.gguf) \| f16 \| 18.6 \| 16 bpw, overkill \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
beichenxie/nikeai-test-2	beichenxie	2025-01-20T23:38:30Z	8	0	diffusers	[ "diffusers", "text-to-image", "diffusers-training", "lora", "replicate", "template:sd-lora", "sd3.5-large", "sd3.5", "sd3.5-diffusers", "base_model:stabilityai/stable-diffusion-3.5-large", "base_model:adapter:stabilityai/stable-diffusion-3.5-large", "license:other", "region:us" ]	text-to-image	2025-01-20T23:32:22Z	--- license: other library_name: diffusers tags: - text-to-image - diffusers-training - diffusers - lora - replicate - template:sd-lora - sd3.5-large - sd3.5 - sd3.5-diffusers base_model: stabilityai/stable-diffusion-3.5-large instance_prompt: HILOS, nikeai test widget: [] --- <!-- This model card has been generated automatically according to the information the training script had access to. You should probably proofread and complete it, then remove this comment. --> # SD3.5-Large DreamBooth LoRA - beichenxie/nikeai-test-2 <Gallery /> ## Model description These are beichenxie/nikeai-test-2 DreamBooth LoRA weights for stable-diffusion-3.5-large. The weights were trained using [DreamBooth](https://dreambooth.github.io/) with the [SD3 diffusers trainer](https://github.com/huggingface/diffusers/blob/main/examples/dreambooth/README_sd3.md). Was LoRA for the text encoder enabled? False. ## Trigger words You should use `HILOS, nikeai test` to trigger the image generation. ## Download model [Download the .safetensors LoRA](beichenxie/nikeai-test-2/tree/main) in the Files & versions tab. ## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers) ```py from diffusers import AutoPipelineForText2Image import torch pipeline = AutoPipelineForText2Image.from_pretrained(stable-diffusion-3.5-large, torch_dtype=torch.float16).to('cuda') pipeline.load_lora_weights('beichenxie/nikeai-test-2', weight_name='pytorch_lora_weights.safetensors') image = pipeline('HILOS, nikeai test').images[0] ``` ### Use it with UIs such as AUTOMATIC1111, Comfy UI, SD.Next, Invoke - LoRA: download [`diffusers_lora_weights.safetensors` here 💾](/beichenxie/nikeai-test-2/blob/main/diffusers_lora_weights.safetensors)*. - Rename it and place it on your `models/Lora` folder. - On AUTOMATIC1111, load the LoRA by adding `<lora:your_new_name:1>` to your prompt. On ComfyUI just [load it as a regular LoRA](https://comfyanonymous.github.io/ComfyUI_examples/lora/). For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters) ## License Please adhere to the licensing terms as described [here](https://huggingface.co/stabilityai/stable-diffusion-3.5-large/blob/main/LICENSE.md). ## Training details Trained on Replicate using: [lucataco/stable-diffusion-3.5-large-lora-trainer](https://replicate.com/lucataco/stable-diffusion-3.5-large-lora-trainer) ## Intended uses & limitations #### How to use ```python # TODO: add an example code snippet for running this diffusion pipeline ``` #### Limitations and bias [TODO: provide examples of latent issues and potential remediations] ## Training details [TODO: describe the data used to train the model]
leixa/d711f2b7-a885-4745-987c-446462155057	leixa	2025-01-20T23:38:29Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:HuggingFaceM4/tiny-random-LlamaForCausalLM", "base_model:adapter:HuggingFaceM4/tiny-random-LlamaForCausalLM", "region:us" ]	null	2025-01-20T23:37:59Z	--- library_name: peft base_model: HuggingFaceM4/tiny-random-LlamaForCausalLM tags: - axolotl - generated_from_trainer model-index: - name: d711f2b7-a885-4745-987c-446462155057 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: HuggingFaceM4/tiny-random-LlamaForCausalLM bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 923c6e5b442f353f_train_data.json ds_type: json format: custom path: /workspace/input_data/923c6e5b442f353f_train_data.json type: field_input: confidence field_instruction: report field_output: statement format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: leixa/d711f2b7-a885-4745-987c-446462155057 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: 0 logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 100 micro_batch_size: 8 mlflow_experiment_name: /tmp/923c6e5b442f353f_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: techspear-hub wandb_mode: online wandb_name: 531f15f8-749c-479d-84e5-335387dc7e76 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 531f15f8-749c-479d-84e5-335387dc7e76 warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # d711f2b7-a885-4745-987c-446462155057 This model is a fine-tuned version of [HuggingFaceM4/tiny-random-LlamaForCausalLM](https://huggingface.co/HuggingFaceM4/tiny-random-LlamaForCausalLM) on the None dataset. It achieves the following results on the evaluation set: - Loss: 10.3697 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 100 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0042 \| 1 \| 10.3767 \| \| 10.3769 \| 0.0379 \| 9 \| 10.3764 \| \| 10.3753 \| 0.0758 \| 18 \| 10.3757 \| \| 10.3743 \| 0.1137 \| 27 \| 10.3749 \| \| 10.3749 \| 0.1516 \| 36 \| 10.3739 \| \| 10.3718 \| 0.1895 \| 45 \| 10.3729 \| \| 10.3726 \| 0.2274 \| 54 \| 10.3719 \| \| 10.3692 \| 0.2653 \| 63 \| 10.3710 \| \| 10.3699 \| 0.3032 \| 72 \| 10.3703 \| \| 10.369 \| 0.3411 \| 81 \| 10.3699 \| \| 10.3684 \| 0.3789 \| 90 \| 10.3698 \| \| 10.3699 \| 0.4168 \| 99 \| 10.3697 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF	mradermacher	2025-01-20T23:38:02Z	573	1	transformers	[ "transformers", "gguf", "text-generation-inference", "unsloth", "gemma2", "trl", "sft", "en", "id", "dataset:gmonsoon/CoT-id", "base_model:gmonsoon/Eunoia-Gemma-9B-o1-Indo", "base_model:quantized:gmonsoon/Eunoia-Gemma-9B-o1-Indo", "license:gemma", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	null	2025-01-17T09:56:31Z	--- base_model: gmonsoon/Eunoia-Gemma-9B-o1-Indo datasets: - gmonsoon/CoT-id language: - en - id library_name: transformers license: gemma quantized_by: mradermacher tags: - text-generation-inference - transformers - unsloth - gemma2 - trl - sft --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: nicoboss --> weighted/imatrix quants of https://huggingface.co/gmonsoon/Eunoia-Gemma-9B-o1-Indo <!-- provided-files --> static quants are available at https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ1_S.gguf) \| i1-IQ1_S \| 2.5 \| for the desperate \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ1_M.gguf) \| i1-IQ1_M \| 2.6 \| mostly desperate \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ2_XXS.gguf) \| i1-IQ2_XXS \| 2.9 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ2_XS.gguf) \| i1-IQ2_XS \| 3.2 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ2_S.gguf) \| i1-IQ2_S \| 3.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ2_M.gguf) \| i1-IQ2_M \| 3.5 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q2_K_S.gguf) \| i1-Q2_K_S \| 3.7 \| very low quality \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ3_XXS.gguf) \| i1-IQ3_XXS \| 3.9 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q2_K.gguf) \| i1-Q2_K \| 3.9 \| IQ3_XXS probably better \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ3_XS.gguf) \| i1-IQ3_XS \| 4.2 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ3_S.gguf) \| i1-IQ3_S \| 4.4 \| beats Q3_K* \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q3_K_S.gguf) \| i1-Q3_K_S \| 4.4 \| IQ3_XS probably better \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ3_M.gguf) \| i1-IQ3_M \| 4.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q3_K_M.gguf) \| i1-Q3_K_M \| 4.9 \| IQ3_S probably better \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q3_K_L.gguf) \| i1-Q3_K_L \| 5.2 \| IQ3_M probably better \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ4_XS.gguf) \| i1-IQ4_XS \| 5.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ4_NL.gguf) \| i1-IQ4_NL \| 5.5 \| prefer IQ4_XS \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q4_0.gguf) \| i1-Q4_0 \| 5.6 \| fast, low quality \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q4_K_S.gguf) \| i1-Q4_K_S \| 5.6 \| optimal size/speed/quality \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q4_K_M.gguf) \| i1-Q4_K_M \| 5.9 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q4_1.gguf) \| i1-Q4_1 \| 6.1 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q5_K_S.gguf) \| i1-Q5_K_S \| 6.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q5_K_M.gguf) \| i1-Q5_K_M \| 6.7 \| \| \| [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q6_K.gguf) \| i1-Q6_K \| 7.7 \| practically like static Q6_K \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to. <!-- end -->
nathanialhunt/a5b35d19-809b-4474-b3e7-2109f98d9b84	nathanialhunt	2025-01-20T23:38:00Z	8	0	peft	[ "peft", "safetensors", "mixtral", "axolotl", "generated_from_trainer", "base_model:Eurdem/Defne_llama3_2x8B", "base_model:adapter:Eurdem/Defne_llama3_2x8B", "license:llama3", "region:us" ]	null	2025-01-20T22:39:16Z	--- library_name: peft license: llama3 base_model: Eurdem/Defne_llama3_2x8B tags: - axolotl - generated_from_trainer model-index: - name: a5b35d19-809b-4474-b3e7-2109f98d9b84 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Eurdem/Defne_llama3_2x8B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 7e3b096caae7eb1c_train_data.json ds_type: json format: custom path: /workspace/input_data/7e3b096caae7eb1c_train_data.json type: field_instruction: instruction field_output: output format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: nathanialhunt/a5b35d19-809b-4474-b3e7-2109f98d9b84 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/7e3b096caae7eb1c_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 319fd169-6b6a-48ba-95dd-c64f160d3657 wandb_project: Birthday-SN56-24-Gradients-On-Demand wandb_run: your_name wandb_runid: 319fd169-6b6a-48ba-95dd-c64f160d3657 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # a5b35d19-809b-4474-b3e7-2109f98d9b84 This model is a fine-tuned version of [Eurdem/Defne_llama3_2x8B](https://huggingface.co/Eurdem/Defne_llama3_2x8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.0000 \| 1 \| nan \| \| 0.0 \| 0.0001 \| 3 \| nan \| \| 0.0 \| 0.0002 \| 6 \| nan \| \| 0.0 \| 0.0002 \| 9 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
lesso10/a4fd6254-1a96-4603-aea9-5a2d20a2f3bc	lesso10	2025-01-20T23:37:34Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2.5-Coder-7B-Instruct", "base_model:adapter:unsloth/Qwen2.5-Coder-7B-Instruct", "license:apache-2.0", "4-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:55:41Z	--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2.5-Coder-7B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: a4fd6254-1a96-4603-aea9-5a2d20a2f3bc results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2.5-Coder-7B-Instruct bf16: auto chat_template: llama3 datasets: - data_files: - 5da4fdb4f9d40cf6_train_data.json ds_type: json format: custom path: /workspace/input_data/5da4fdb4f9d40cf6_train_data.json type: field_input: topic; field_instruction: message_1 field_output: message_2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: true gradient_checkpointing: true group_by_length: false hub_model_id: lesso10/a4fd6254-1a96-4603-aea9-5a2d20a2f3bc hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: true load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/5da4fdb4f9d40cf6_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 971c59eb-5de8-4a78-8d22-6a7da4c9ee82 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 971c59eb-5de8-4a78-8d22-6a7da4c9ee82 warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # a4fd6254-1a96-4603-aea9-5a2d20a2f3bc This model is a fine-tuned version of [unsloth/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0001 \| 1 \| nan \| \| 0.0 \| 0.0005 \| 5 \| nan \| \| 0.0 \| 0.0011 \| 10 \| nan \| \| 0.0 \| 0.0016 \| 15 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
LHRuig/detailsharp	LHRuig	2025-01-20T23:37:03Z	28	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:36:29Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # detailsharp <Gallery /> ## Model description detailsharp lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/detailsharp/tree/main) them in the Files & versions tab.
matrixportal/Llama-3.1-8B-BookAdventures-GGUF	matrixportal	2025-01-20T23:36:24Z	28	0	null	[ "gguf", "llama-factory", "full", "generated_from_trainer", "llama-cpp", "gguf-my-repo", "base_model:KoboldAI/Llama-3.1-8B-BookAdventures", "base_model:quantized:KoboldAI/Llama-3.1-8B-BookAdventures", "license:cc-by-nc-sa-4.0", "endpoints_compatible", "region:us", "conversational" ]	null	2025-01-20T23:35:43Z	--- license: cc-by-nc-sa-4.0 base_model: KoboldAI/Llama-3.1-8B-BookAdventures tags: - llama-factory - full - generated_from_trainer - llama-cpp - gguf-my-repo model-index: - name: KoboldAI/Llama-3.1-8B-BookAdventures results: [] --- # matrixportal/Llama-3.1-8B-BookAdventures-GGUF This model was converted to GGUF format from [`KoboldAI/Llama-3.1-8B-BookAdventures`](https://huggingface.co/KoboldAI/Llama-3.1-8B-BookAdventures) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space. Refer to the [original model card](https://huggingface.co/KoboldAI/Llama-3.1-8B-BookAdventures) for more details on the model. ## Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) ```bash brew install llama.cpp ``` Invoke the llama.cpp server or the CLI. ### CLI: ```bash llama-cli --hf-repo matrixportal/Llama-3.1-8B-BookAdventures-GGUF --hf-file llama-3.1-8b-bookadventures-q4_k_m.gguf -p "The meaning to life and the universe is" ``` ### Server: ```bash llama-server --hf-repo matrixportal/Llama-3.1-8B-BookAdventures-GGUF --hf-file llama-3.1-8b-bookadventures-q4_k_m.gguf -c 2048 ``` Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well. Step 1: Clone llama.cpp from GitHub. ``` git clone https://github.com/ggerganov/llama.cpp ``` Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux). ``` cd llama.cpp && LLAMA_CURL=1 make ``` Step 3: Run inference through the main binary. ``` ./llama-cli --hf-repo matrixportal/Llama-3.1-8B-BookAdventures-GGUF --hf-file llama-3.1-8b-bookadventures-q4_k_m.gguf -p "The meaning to life and the universe is" ``` or ``` ./llama-server --hf-repo matrixportal/Llama-3.1-8B-BookAdventures-GGUF --hf-file llama-3.1-8b-bookadventures-q4_k_m.gguf -c 2048 ```
fpadovani/english_childes_context_42	fpadovani	2025-01-20T23:36:06Z	6	0	transformers	[ "transformers", "safetensors", "roberta", "fill-mask", "generated_from_trainer", "autotrain_compatible", "endpoints_compatible", "region:us" ]	fill-mask	2025-01-20T22:20:37Z	--- library_name: transformers tags: - generated_from_trainer model-index: - name: childes_mlm_unmasking_context results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # childes_mlm_unmasking_context This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 2.1318 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 100000 - training_steps: 400000 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:-------:\|:-----:\|:---------------:\| \| No log \| 1.2698 \| 2000 \| 5.5274 \| \| 6.237 \| 2.5397 \| 4000 \| 5.4702 \| \| 6.237 \| 3.8095 \| 6000 \| 5.3851 \| \| 5.4081 \| 5.0794 \| 8000 \| 5.3058 \| \| 5.4081 \| 6.3492 \| 10000 \| 4.0259 \| \| 4.2991 \| 7.6190 \| 12000 \| 3.3054 \| \| 4.2991 \| 8.8889 \| 14000 \| 3.0169 \| \| 3.104 \| 10.1587 \| 16000 \| 2.7758 \| \| 3.104 \| 11.4286 \| 18000 \| 2.6658 \| \| 2.7387 \| 12.6984 \| 20000 \| 2.5578 \| \| 2.7387 \| 13.9683 \| 22000 \| 2.5101 \| \| 2.5352 \| 15.2381 \| 24000 \| 2.4254 \| \| 2.5352 \| 16.5079 \| 26000 \| 2.3459 \| \| 2.4005 \| 17.7778 \| 28000 \| 2.3221 \| \| 2.4005 \| 19.0476 \| 30000 \| 2.2965 \| \| 2.3081 \| 20.3175 \| 32000 \| 2.2853 \| \| 2.3081 \| 21.5873 \| 34000 \| 2.2284 \| \| 2.2426 \| 22.8571 \| 36000 \| 2.1734 \| \| 2.2426 \| 24.1270 \| 38000 \| 2.1705 \| \| 2.193 \| 25.3968 \| 40000 \| 2.1931 \| \| 2.193 \| 26.6667 \| 42000 \| 2.1680 \| \| 2.1678 \| 27.9365 \| 44000 \| 2.1337 \| \| 2.1678 \| 29.2063 \| 46000 \| 2.1411 \| \| 2.1464 \| 30.4762 \| 48000 \| 2.1310 \| \| 2.1464 \| 31.7460 \| 50000 \| 2.1229 \| \| 2.1285 \| 33.0159 \| 52000 \| 2.1331 \| \| 2.1285 \| 34.2857 \| 54000 \| 2.1592 \| \| 2.1199 \| 35.5556 \| 56000 \| 2.1318 \| ### Framework versions - Transformers 4.45.2 - Pytorch 2.5.1+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
dimasik1987/d0b444d3-4c6e-4c6f-a3e0-c745e20ca3f8	dimasik1987	2025-01-20T23:36:04Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:trl-internal-testing/tiny-random-LlamaForCausalLM", "base_model:adapter:trl-internal-testing/tiny-random-LlamaForCausalLM", "region:us" ]	null	2025-01-20T23:35:41Z	--- library_name: peft base_model: trl-internal-testing/tiny-random-LlamaForCausalLM tags: - axolotl - generated_from_trainer model-index: - name: d0b444d3-4c6e-4c6f-a3e0-c745e20ca3f8 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: trl-internal-testing/tiny-random-LlamaForCausalLM bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - dd22c8863ed4176b_train_data.json ds_type: json format: custom path: /workspace/input_data/dd22c8863ed4176b_train_data.json type: field_input: text field_instruction: title field_output: summary format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: dimasik1987/d0b444d3-4c6e-4c6f-a3e0-c745e20ca3f8 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/dd22c8863ed4176b_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ee0747e5-378f-43ac-83d3-8dd08d6876bf wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ee0747e5-378f-43ac-83d3-8dd08d6876bf warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # d0b444d3-4c6e-4c6f-a3e0-c745e20ca3f8 This model is a fine-tuned version of [trl-internal-testing/tiny-random-LlamaForCausalLM](https://huggingface.co/trl-internal-testing/tiny-random-LlamaForCausalLM) on the None dataset. It achieves the following results on the evaluation set: - Loss: 10.3303 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0292 \| 1 \| 10.3601 \| \| 10.3594 \| 0.1460 \| 5 \| 10.3562 \| \| 10.3515 \| 0.2920 \| 10 \| 10.3464 \| \| 10.3409 \| 0.4380 \| 15 \| 10.3383 \| \| 10.337 \| 0.5839 \| 20 \| 10.3330 \| \| 10.3318 \| 0.7299 \| 25 \| 10.3307 \| \| 10.3292 \| 0.8759 \| 30 \| 10.3303 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
LHRuig/moodcine	LHRuig	2025-01-20T23:35:59Z	12	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:35:47Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # moodcine <Gallery /> ## Model description moodcine lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/moodcine/tree/main) them in the Files & versions tab.
prxy5604/37578cc4-aa06-437b-80e1-561bf03536ef	prxy5604	2025-01-20T23:35:57Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:trl-internal-testing/tiny-random-LlamaForCausalLM", "base_model:adapter:trl-internal-testing/tiny-random-LlamaForCausalLM", "region:us" ]	null	2025-01-20T23:35:36Z	--- library_name: peft base_model: trl-internal-testing/tiny-random-LlamaForCausalLM tags: - axolotl - generated_from_trainer model-index: - name: 37578cc4-aa06-437b-80e1-561bf03536ef results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: trl-internal-testing/tiny-random-LlamaForCausalLM bf16: true chat_template: llama3 data_processes: 16 dataset_prepared_path: null datasets: - data_files: - dd22c8863ed4176b_train_data.json ds_type: json format: custom path: /workspace/input_data/dd22c8863ed4176b_train_data.json type: field_input: text field_instruction: title field_output: summary format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto do_eval: true early_stopping_patience: 5 eval_batch_size: 4 eval_max_new_tokens: 128 eval_steps: 50 eval_table_size: null evals_per_epoch: null flash_attention: true fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: prxy5604/37578cc4-aa06-437b-80e1-561bf03536ef hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 128 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 64 lora_target_linear: true lr_scheduler: cosine max_grad_norm: 1.0 max_memory: 0: 75GB max_steps: 200 micro_batch_size: 8 mlflow_experiment_name: /tmp/dd22c8863ed4176b_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 50 saves_per_epoch: null sequence_len: 1024 strict: false tf32: true tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ee0747e5-378f-43ac-83d3-8dd08d6876bf wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ee0747e5-378f-43ac-83d3-8dd08d6876bf warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 37578cc4-aa06-437b-80e1-561bf03536ef This model is a fine-tuned version of [trl-internal-testing/tiny-random-LlamaForCausalLM](https://huggingface.co/trl-internal-testing/tiny-random-LlamaForCausalLM) on the None dataset. It achieves the following results on the evaluation set: - Loss: 10.2951 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 8 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 52 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 10.3598 \| 0.0580 \| 1 \| 10.3596 \| \| 10.5029 \| 2.8986 \| 50 \| 10.2951 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
JacksonBrune/919b9850-bcff-4436-97bc-01c41a6c1517	JacksonBrune	2025-01-20T23:35:13Z	6	0	peft	[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "base_model:Intel/neural-chat-7b-v3-3", "base_model:adapter:Intel/neural-chat-7b-v3-3", "license:apache-2.0", "region:us" ]	null	2025-01-20T20:14:32Z	--- library_name: peft license: apache-2.0 base_model: Intel/neural-chat-7b-v3-3 tags: - axolotl - generated_from_trainer model-index: - name: 919b9850-bcff-4436-97bc-01c41a6c1517 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Intel/neural-chat-7b-v3-3 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 50587f38ed52161f_train_data.json ds_type: json format: custom path: /workspace/input_data/50587f38ed52161f_train_data.json type: field_input: file field_instruction: directory field_output: content format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: JacksonBrune/919b9850-bcff-4436-97bc-01c41a6c1517 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/50587f38ed52161f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 3f448567-d842-4770-a570-924da07f2a6c wandb_project: Birthday-SN56-12-Gradients-On-Demand wandb_run: your_name wandb_runid: 3f448567-d842-4770-a570-924da07f2a6c warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 919b9850-bcff-4436-97bc-01c41a6c1517 This model is a fine-tuned version of [Intel/neural-chat-7b-v3-3](https://huggingface.co/Intel/neural-chat-7b-v3-3) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.0000 \| 1 \| nan \| \| 0.0 \| 0.0000 \| 3 \| nan \| \| 0.0 \| 0.0000 \| 6 \| nan \| \| 0.0 \| 0.0000 \| 9 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
mradermacher/NBeerbower-ConversationalMix-8b-GGUF	mradermacher	2025-01-20T23:34:47Z	328	1	transformers	[ "transformers", "gguf", "mergekit", "merge", "en", "base_model:MrRobotoAI/NBeerbower-ConversationalMix-8b", "base_model:quantized:MrRobotoAI/NBeerbower-ConversationalMix-8b", "endpoints_compatible", "region:us", "conversational" ]	null	2025-01-19T21:42:40Z	--- base_model: MrRobotoAI/NBeerbower-ConversationalMix-8b language: - en library_name: transformers quantized_by: mradermacher tags: - mergekit - merge --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/MrRobotoAI/NBeerbower-ConversationalMix-8b <!-- provided-files --> weighted/imatrix quants are available at https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-i1-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q2_K.gguf) \| Q2_K \| 3.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q3_K_S.gguf) \| Q3_K_S \| 3.8 \| \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q3_K_M.gguf) \| Q3_K_M \| 4.1 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q3_K_L.gguf) \| Q3_K_L \| 4.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.IQ4_XS.gguf) \| IQ4_XS \| 4.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q4_K_S.gguf) \| Q4_K_S \| 4.8 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q4_K_M.gguf) \| Q4_K_M \| 5.0 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q5_K_S.gguf) \| Q5_K_S \| 5.7 \| \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q5_K_M.gguf) \| Q5_K_M \| 5.8 \| \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q6_K.gguf) \| Q6_K \| 6.7 \| very good quality \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q8_0.gguf) \| Q8_0 \| 8.6 \| fast, best quality \| \| [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.f16.gguf) \| f16 \| 16.2 \| 16 bpw, overkill \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
lesso13/d9c12229-8dbb-45fa-ba4f-9717c95db112	lesso13	2025-01-20T23:34:35Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "base_model:adapter:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:03:26Z	--- library_name: peft base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f tags: - axolotl - generated_from_trainer model-index: - name: d9c12229-8dbb-45fa-ba4f-9717c95db112 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f bf16: true chat_template: llama3 datasets: - data_files: - train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json ds_type: json format: custom path: /workspace/input_data/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso13/d9c12229-8dbb-45fa-ba4f-9717c95db112 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 45GiB max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075526eb-32e0-4485-aab7-014e4d302171 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075526eb-32e0-4485-aab7-014e4d302171 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # d9c12229-8dbb-45fa-ba4f-9717c95db112 This model is a fine-tuned version of [samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f](https://huggingface.co/samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.0002 \| 1 \| nan \| \| 0.0 \| 0.0010 \| 5 \| nan \| \| 0.0 \| 0.0020 \| 10 \| nan \| \| 0.0 \| 0.0031 \| 15 \| nan \| \| 0.0 \| 0.0041 \| 20 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
demohong/7f82ae83-b307-42ac-b3a7-914fcf126608	demohong	2025-01-20T23:30:24Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:unsloth/Meta-Llama-3.1-8B", "base_model:adapter:unsloth/Meta-Llama-3.1-8B", "license:llama3.1", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:59:26Z	--- library_name: peft license: llama3.1 base_model: unsloth/Meta-Llama-3.1-8B tags: - axolotl - generated_from_trainer model-index: - name: 7f82ae83-b307-42ac-b3a7-914fcf126608 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Meta-Llama-3.1-8B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 91e193d3dca1611f_train_data.json ds_type: json format: custom path: /workspace/input_data/91e193d3dca1611f_train_data.json type: field_input: parent_id field_instruction: role field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: demohong/7f82ae83-b307-42ac-b3a7-914fcf126608 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/91e193d3dca1611f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 856a9aac-189f-40f7-b27c-c5616995b0d1 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 856a9aac-189f-40f7-b27c-c5616995b0d1 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 7f82ae83-b307-42ac-b3a7-914fcf126608 This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.5966 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.3105 \| 0.0705 \| 200 \| 1.5966 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
nbninh/d2c84520-d1af-4aa3-b9a1-d0cdf04d4d4d	nbninh	2025-01-20T23:29:25Z	8	0	peft	[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "base_model:unsloth/Phi-3-medium-4k-instruct", "base_model:adapter:unsloth/Phi-3-medium-4k-instruct", "license:mit", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T21:45:32Z	--- library_name: peft license: mit base_model: unsloth/Phi-3-medium-4k-instruct tags: - axolotl - generated_from_trainer model-index: - name: d2c84520-d1af-4aa3-b9a1-d0cdf04d4d4d results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Phi-3-medium-4k-instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 41d403c8b37c92fc_train_data.json ds_type: json format: custom path: /workspace/input_data/41d403c8b37c92fc_train_data.json type: field_input: mesh_terms field_instruction: title field_output: abstract format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nbninh/d2c84520-d1af-4aa3-b9a1-d0cdf04d4d4d hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/41d403c8b37c92fc_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075ea541-bd04-429e-a989-c49dabc36fc3 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075ea541-bd04-429e-a989-c49dabc36fc3 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # d2c84520-d1af-4aa3-b9a1-d0cdf04d4d4d This model is a fine-tuned version of [unsloth/Phi-3-medium-4k-instruct](https://huggingface.co/unsloth/Phi-3-medium-4k-instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.4790 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 5.7635 \| 0.0048 \| 200 \| 1.4790 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
LHRuig/detail2k	LHRuig	2025-01-20T23:29:11Z	5	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:28:58Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # detail2k <Gallery /> ## Model description detail2k lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/detail2k/tree/main) them in the Files & versions tab.
MikeRoz/deepseek-ai_DeepSeek-R1-Distill-Llama-70B-4.25bpw-h6-exl2	MikeRoz	2025-01-20T23:28:11Z	227	4	null	[ "safetensors", "llama", "exl2", "region:us" ]	null	2025-01-20T20:50:11Z	# DeepSeek-R1 <!-- markdownlint-disable first-line-h1 --> <!-- markdownlint-disable html --> <!-- markdownlint-disable no-duplicate-header --> <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- Post-Training: Large-Scale Reinforcement Learning on the Base Model - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- Distillation: Smaller Models Can Be Powerful Too - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> \| Model \| #Total Params \| #Activated Params \| Context Length \| Download \| \| :------------: \| :------------: \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Zero \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) \| \| DeepSeek-R1 \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) \| </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> \| Model \| Base Model \| Download \| \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Distill-Qwen-1.5B \| [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) \| \| DeepSeek-R1-Distill-Qwen-7B \| [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) \| \| DeepSeek-R1-Distill-Llama-8B \| [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) \| \| DeepSeek-R1-Distill-Qwen-14B \| [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) \| \|DeepSeek-R1-Distill-Qwen-32B \| [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) \| \| DeepSeek-R1-Distill-Llama-70B \| [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) \| </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> \| Category \| Benchmark (Metric) \| Claude-3.5-Sonnet-1022 \| GPT-4o 0513 \| DeepSeek V3 \| OpenAI o1-mini \| OpenAI o1-1217 \| DeepSeek R1 \| \|----------\|-------------------\|----------------------\|------------\|--------------\|----------------\|------------\|--------------\| \| \| Architecture \| - \| - \| MoE \| - \| - \| MoE \| \| \| # Activated Params \| - \| - \| 37B \| - \| - \| 37B \| \| \| # Total Params \| - \| - \| 671B \| - \| - \| 671B \| \| English \| MMLU (Pass@1) \| 88.3 \| 87.2 \| 88.5 \| 85.2 \| 91.8 \| 90.8 \| \| \| MMLU-Redux (EM) \| 88.9 \| 88.0 \| 89.1 \| 86.7 \| - \| 92.9 \| \| \| MMLU-Pro (EM) \| 78.0 \| 72.6 \| 75.9 \| 80.3 \| - \| 84.0 \| \| \| DROP (3-shot F1) \| 88.3 \| 83.7 \| 91.6 \| 83.9 \| 90.2 \| 92.2 \| \| \| IF-Eval (Prompt Strict) \| 86.5 \| 84.3 \| 86.1 \| 84.8 \| - \| 83.3 \| \| \| GPQA-Diamond (Pass@1) \| 65.0 \| 49.9 \| 59.1 \| 60.0 \| 75.7 \| 71.5 \| \| \| SimpleQA (Correct) \| 28.4 \| 38.2 \| 24.9 \| 7.0 \| 47.0 \| 30.1 \| \| \| FRAMES (Acc.) \| 72.5 \| 80.5 \| 73.3 \| 76.9 \| - \| 82.5 \| \| \| AlpacaEval2.0 (LC-winrate) \| 52.0 \| 51.1 \| 70.0 \| 57.8 \| - \| 87.6 \| \| \| ArenaHard (GPT-4-1106) \| 85.2 \| 80.4 \| 85.5 \| 92.0 \| - \| 92.3 \| \| Code \| LiveCodeBench (Pass@1-COT) \| 33.8 \| 34.2 \| - \| 53.8 \| 63.4 \| 65.9 \| \| \| Codeforces (Percentile) \| 20.3 \| 23.6 \| 58.7 \| 93.4 \| 96.6 \| 96.3 \| \| \| Codeforces (Rating) \| 717 \| 759 \| 1134 \| 1820 \| 2061 \| 2029 \| \| \| SWE Verified (Resolved) \| 50.8 \| 38.8 \| 42.0 \| 41.6 \| 48.9 \| 49.2 \| \| \| Aider-Polyglot (Acc.) \| 45.3 \| 16.0 \| 49.6 \| 32.9 \| 61.7 \| 53.3 \| \| Math \| AIME 2024 (Pass@1) \| 16.0 \| 9.3 \| 39.2 \| 63.6 \| 79.2 \| 79.8 \| \| \| MATH-500 (Pass@1) \| 78.3 \| 74.6 \| 90.2 \| 90.0 \| 96.4 \| 97.3 \| \| \| CNMO 2024 (Pass@1) \| 13.1 \| 10.8 \| 43.2 \| 67.6 \| - \| 78.8 \| \| Chinese \| CLUEWSC (EM) \| 85.4 \| 87.9 \| 90.9 \| 89.9 \| - \| 92.8 \| \| \| C-Eval (EM) \| 76.7 \| 76.0 \| 86.5 \| 68.9 \| - \| 91.8 \| \| \| C-SimpleQA (Correct) \| 55.4 \| 58.7 \| 68.0 \| 40.3 \| - \| 63.7 \| </div> ### Distilled Model Evaluation <div align="center"> \| Model \| AIME 2024 pass@1 \| AIME 2024 cons@64 \| MATH-500 pass@1 \| GPQA Diamond pass@1 \| LiveCodeBench pass@1 \| CodeForces rating \| \|------------------------------------------\|------------------\|-------------------\|-----------------\|----------------------\|----------------------\|-------------------\| \| GPT-4o-0513 \| 9.3 \| 13.4 \| 74.6 \| 49.9 \| 32.9 \| 759 \| \| Claude-3.5-Sonnet-1022 \| 16.0 \| 26.7 \| 78.3 \| 65.0 \| 38.9 \| 717 \| \| o1-mini \| 63.6 \| 80.0 \| 90.0 \| 60.0 \| 53.8 \| 1820 \| \| QwQ-32B-Preview \| 44.0 \| 60.0 \| 90.6 \| 54.5 \| 41.9 \| 1316 \| \| DeepSeek-R1-Distill-Qwen-1.5B \| 28.9 \| 52.7 \| 83.9 \| 33.8 \| 16.9 \| 954 \| \| DeepSeek-R1-Distill-Qwen-7B \| 55.5 \| 83.3 \| 92.8 \| 49.1 \| 37.6 \| 1189 \| \| DeepSeek-R1-Distill-Qwen-14B \| 69.7 \| 80.0 \| 93.9 \| 59.1 \| 53.1 \| 1481 \| \| DeepSeek-R1-Distill-Qwen-32B \| 72.6 \| 83.3 \| 94.3 \| 62.1 \| 57.2 \| 1691 \| \| DeepSeek-R1-Distill-Llama-8B \| 50.4 \| 80.0 \| 89.1 \| 49.0 \| 39.6 \| 1205 \| \| DeepSeek-R1-Distill-Llama-70B \| 70.0 \| 86.7 \| 94.5 \| 65.2 \| 57.5 \| 1633 \| </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output. ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).
LHRuig/adddetails	LHRuig	2025-01-20T23:26:11Z	9	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:26:06Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # adddetail <Gallery /> ## Model description adddetail lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/adddetails/tree/main) them in the Files & versions tab.
lesso10/1c6a64b4-e7f1-4e45-9efb-b4b0d937a28a	lesso10	2025-01-20T23:25:07Z	8	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/CodeLlama-13b-hf-flash", "base_model:adapter:NousResearch/CodeLlama-13b-hf-flash", "4-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:41:43Z	--- library_name: peft base_model: NousResearch/CodeLlama-13b-hf-flash tags: - axolotl - generated_from_trainer model-index: - name: 1c6a64b4-e7f1-4e45-9efb-b4b0d937a28a results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/CodeLlama-13b-hf-flash bf16: auto chat_template: llama3 datasets: - data_files: - 9ff4e3b24bf3b2a4_train_data.json ds_type: json format: custom path: /workspace/input_data/9ff4e3b24bf3b2a4_train_data.json type: field_input: sentence1 field_instruction: phrase1 field_output: sentence2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: true gradient_checkpointing: true group_by_length: false hub_model_id: lesso10/1c6a64b4-e7f1-4e45-9efb-b4b0d937a28a hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: true load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/9ff4e3b24bf3b2a4_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # 1c6a64b4-e7f1-4e45-9efb-b4b0d937a28a This model is a fine-tuned version of [NousResearch/CodeLlama-13b-hf-flash](https://huggingface.co/NousResearch/CodeLlama-13b-hf-flash) on the None dataset. It achieves the following results on the evaluation set: - Loss: 3.0292 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0003 \| 1 \| 4.9637 \| \| 5.0443 \| 0.0014 \| 5 \| 4.8203 \| \| 4.2709 \| 0.0029 \| 10 \| 3.9057 \| \| 3.0327 \| 0.0043 \| 15 \| 3.1180 \| \| 3.3112 \| 0.0057 \| 20 \| 3.0652 \| \| 2.8835 \| 0.0071 \| 25 \| 3.0292 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
John6666/rippillustrious-v10-sdxl	John6666	2025-01-20T23:24:26Z	81	0	diffusers	[ "diffusers", "safetensors", "text-to-image", "stable-diffusion", "stable-diffusion-xl", "anime", "girls", "illustrious", "en", "base_model:OnomaAIResearch/Illustrious-xl-early-release-v0", "base_model:finetune:OnomaAIResearch/Illustrious-xl-early-release-v0", "license:other", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionXLPipeline", "region:us" ]	text-to-image	2025-01-20T23:18:28Z	--- license: other license_name: faipl-1.0-sd license_link: https://freedevproject.org/faipl-1.0-sd/ language: - en library_name: diffusers pipeline_tag: text-to-image tags: - text-to-image - stable-diffusion - stable-diffusion-xl - anime - girls - illustrious base_model: OnomaAIResearch/Illustrious-xl-early-release-v0 --- Original model is [here](https://civitai.com/models/1163094/rippillustrious?modelVersionId=1308356). This model created by [tbets182132](https://civitai.com/user/tbets182132).
LHRuig/kodakmotion	LHRuig	2025-01-20T23:24:01Z	11	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:23:17Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # kodakmotion <Gallery /> ## Model description kodakmotion lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/kodakmotion/tree/main) them in the Files & versions tab.
lhong4759/bbef3a84-7d4f-465d-914c-e5286aa7e060	lhong4759	2025-01-20T23:23:56Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "base_model:adapter:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:03:00Z	--- library_name: peft base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f tags: - axolotl - generated_from_trainer model-index: - name: bbef3a84-7d4f-465d-914c-e5286aa7e060 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json ds_type: json format: custom path: /workspace/input_data/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: lhong4759/bbef3a84-7d4f-465d-914c-e5286aa7e060 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075526eb-32e0-4485-aab7-014e4d302171 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075526eb-32e0-4485-aab7-014e4d302171 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # bbef3a84-7d4f-465d-914c-e5286aa7e060 This model is a fine-tuned version of [samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f](https://huggingface.co/samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.1312 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.2357 \| 0.0407 \| 200 \| 1.1312 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
adamo1139/DeepSeek-R1-Distill-Qwen-1.5B-3bpw-exl2	adamo1139	2025-01-20T23:23:37Z	5	0	null	[ "qwen2", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "3-bit", "exl2", "region:us" ]	null	2025-01-20T22:53:24Z	--- base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --- # DeepSeek-R1 <!-- markdownlint-disable first-line-h1 --> <!-- markdownlint-disable html --> <!-- markdownlint-disable no-duplicate-header --> <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- Post-Training: Large-Scale Reinforcement Learning on the Base Model - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- Distillation: Smaller Models Can Be Powerful Too - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> \| Model \| #Total Params \| #Activated Params \| Context Length \| Download \| \| :------------: \| :------------: \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Zero \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) \| \| DeepSeek-R1 \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) \| </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> \| Model \| Base Model \| Download \| \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Distill-Qwen-1.5B \| [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) \| \| DeepSeek-R1-Distill-Qwen-7B \| [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) \| \| DeepSeek-R1-Distill-Llama-8B \| [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) \| \| DeepSeek-R1-Distill-Qwen-14B \| [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) \| \|DeepSeek-R1-Distill-Qwen-32B \| [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) \| \| DeepSeek-R1-Distill-Llama-70B \| [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) \| </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> \| Category \| Benchmark (Metric) \| Claude-3.5-Sonnet-1022 \| GPT-4o 0513 \| DeepSeek V3 \| OpenAI o1-mini \| OpenAI o1-1217 \| DeepSeek R1 \| \|----------\|-------------------\|----------------------\|------------\|--------------\|----------------\|------------\|--------------\| \| \| Architecture \| - \| - \| MoE \| - \| - \| MoE \| \| \| # Activated Params \| - \| - \| 37B \| - \| - \| 37B \| \| \| # Total Params \| - \| - \| 671B \| - \| - \| 671B \| \| English \| MMLU (Pass@1) \| 88.3 \| 87.2 \| 88.5 \| 85.2 \| 91.8 \| 90.8 \| \| \| MMLU-Redux (EM) \| 88.9 \| 88.0 \| 89.1 \| 86.7 \| - \| 92.9 \| \| \| MMLU-Pro (EM) \| 78.0 \| 72.6 \| 75.9 \| 80.3 \| - \| 84.0 \| \| \| DROP (3-shot F1) \| 88.3 \| 83.7 \| 91.6 \| 83.9 \| 90.2 \| 92.2 \| \| \| IF-Eval (Prompt Strict) \| 86.5 \| 84.3 \| 86.1 \| 84.8 \| - \| 83.3 \| \| \| GPQA-Diamond (Pass@1) \| 65.0 \| 49.9 \| 59.1 \| 60.0 \| 75.7 \| 71.5 \| \| \| SimpleQA (Correct) \| 28.4 \| 38.2 \| 24.9 \| 7.0 \| 47.0 \| 30.1 \| \| \| FRAMES (Acc.) \| 72.5 \| 80.5 \| 73.3 \| 76.9 \| - \| 82.5 \| \| \| AlpacaEval2.0 (LC-winrate) \| 52.0 \| 51.1 \| 70.0 \| 57.8 \| - \| 87.6 \| \| \| ArenaHard (GPT-4-1106) \| 85.2 \| 80.4 \| 85.5 \| 92.0 \| - \| 92.3 \| \| Code \| LiveCodeBench (Pass@1-COT) \| 33.8 \| 34.2 \| - \| 53.8 \| 63.4 \| 65.9 \| \| \| Codeforces (Percentile) \| 20.3 \| 23.6 \| 58.7 \| 93.4 \| 96.6 \| 96.3 \| \| \| Codeforces (Rating) \| 717 \| 759 \| 1134 \| 1820 \| 2061 \| 2029 \| \| \| SWE Verified (Resolved) \| 50.8 \| 38.8 \| 42.0 \| 41.6 \| 48.9 \| 49.2 \| \| \| Aider-Polyglot (Acc.) \| 45.3 \| 16.0 \| 49.6 \| 32.9 \| 61.7 \| 53.3 \| \| Math \| AIME 2024 (Pass@1) \| 16.0 \| 9.3 \| 39.2 \| 63.6 \| 79.2 \| 79.8 \| \| \| MATH-500 (Pass@1) \| 78.3 \| 74.6 \| 90.2 \| 90.0 \| 96.4 \| 97.3 \| \| \| CNMO 2024 (Pass@1) \| 13.1 \| 10.8 \| 43.2 \| 67.6 \| - \| 78.8 \| \| Chinese \| CLUEWSC (EM) \| 85.4 \| 87.9 \| 90.9 \| 89.9 \| - \| 92.8 \| \| \| C-Eval (EM) \| 76.7 \| 76.0 \| 86.5 \| 68.9 \| - \| 91.8 \| \| \| C-SimpleQA (Correct) \| 55.4 \| 58.7 \| 68.0 \| 40.3 \| - \| 63.7 \| </div> ### Distilled Model Evaluation <div align="center"> \| Model \| AIME 2024 pass@1 \| AIME 2024 cons@64 \| MATH-500 pass@1 \| GPQA Diamond pass@1 \| LiveCodeBench pass@1 \| CodeForces rating \| \|------------------------------------------\|------------------\|-------------------\|-----------------\|----------------------\|----------------------\|-------------------\| \| GPT-4o-0513 \| 9.3 \| 13.4 \| 74.6 \| 49.9 \| 32.9 \| 759 \| \| Claude-3.5-Sonnet-1022 \| 16.0 \| 26.7 \| 78.3 \| 65.0 \| 38.9 \| 717 \| \| o1-mini \| 63.6 \| 80.0 \| 90.0 \| 60.0 \| 53.8 \| 1820 \| \| QwQ-32B-Preview \| 44.0 \| 60.0 \| 90.6 \| 54.5 \| 41.9 \| 1316 \| \| DeepSeek-R1-Distill-Qwen-1.5B \| 28.9 \| 52.7 \| 83.9 \| 33.8 \| 16.9 \| 954 \| \| DeepSeek-R1-Distill-Qwen-7B \| 55.5 \| 83.3 \| 92.8 \| 49.1 \| 37.6 \| 1189 \| \| DeepSeek-R1-Distill-Qwen-14B \| 69.7 \| 80.0 \| 93.9 \| 59.1 \| 53.1 \| 1481 \| \| DeepSeek-R1-Distill-Qwen-32B \| 72.6 \| 83.3 \| 94.3 \| 62.1 \| 57.2 \| 1691 \| \| DeepSeek-R1-Distill-Llama-8B \| 50.4 \| 80.0 \| 89.1 \| 49.0 \| 39.6 \| 1205 \| \| DeepSeek-R1-Distill-Llama-70B \| 70.0 \| 86.7 \| 94.5 \| 65.2 \| 57.5 \| 1633 \| </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output. ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).
mradermacher/LwQ-Reasoner-10B-GGUF	mradermacher	2025-01-20T23:23:24Z	345	0	transformers	[ "transformers", "gguf", "LlamaWithQuestions", "CoT", "Reasoner", "LWQ", "en", "base_model:prithivMLmods/LwQ-Reasoner-10B", "base_model:quantized:prithivMLmods/LwQ-Reasoner-10B", "license:llama3.1", "endpoints_compatible", "region:us" ]	null	2025-01-20T16:50:16Z	--- base_model: prithivMLmods/LwQ-Reasoner-10B language: - en library_name: transformers license: llama3.1 quantized_by: mradermacher tags: - LlamaWithQuestions - CoT - Reasoner - LWQ --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/prithivMLmods/LwQ-Reasoner-10B <!-- provided-files --> weighted/imatrix quants are available at https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q2_K.gguf) \| Q2_K \| 4.0 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q3_K_S.gguf) \| Q3_K_S \| 4.7 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q3_K_M.gguf) \| Q3_K_M \| 5.2 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q3_K_L.gguf) \| Q3_K_L \| 5.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.IQ4_XS.gguf) \| IQ4_XS \| 5.8 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q4_K_S.gguf) \| Q4_K_S \| 6.1 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q4_K_M.gguf) \| Q4_K_M \| 6.4 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q5_K_S.gguf) \| Q5_K_S \| 7.2 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q5_K_M.gguf) \| Q5_K_M \| 7.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q6_K.gguf) \| Q6_K \| 8.6 \| very good quality \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q8_0.gguf) \| Q8_0 \| 11.1 \| fast, best quality \| \| [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.f16.gguf) \| f16 \| 20.7 \| 16 bpw, overkill \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
LHRuig/hollycine	LHRuig	2025-01-20T23:23:02Z	6	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:22:02Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # hollycine <Gallery /> ## Model description hollycine lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/hollycine/tree/main) them in the Files & versions tab.
adamo1139/DeepSeek-R1-Distill-Qwen-1.5B-5bpw-exl2	adamo1139	2025-01-20T23:22:53Z	5	0	null	[ "qwen2", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "5-bit", "exl2", "region:us" ]	null	2025-01-20T22:51:52Z	--- base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --- # DeepSeek-R1 <!-- markdownlint-disable first-line-h1 --> <!-- markdownlint-disable html --> <!-- markdownlint-disable no-duplicate-header --> <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- Post-Training: Large-Scale Reinforcement Learning on the Base Model - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- Distillation: Smaller Models Can Be Powerful Too - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> \| Model \| #Total Params \| #Activated Params \| Context Length \| Download \| \| :------------: \| :------------: \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Zero \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) \| \| DeepSeek-R1 \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) \| </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> \| Model \| Base Model \| Download \| \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Distill-Qwen-1.5B \| [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) \| \| DeepSeek-R1-Distill-Qwen-7B \| [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) \| \| DeepSeek-R1-Distill-Llama-8B \| [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) \| \| DeepSeek-R1-Distill-Qwen-14B \| [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) \| \|DeepSeek-R1-Distill-Qwen-32B \| [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) \| \| DeepSeek-R1-Distill-Llama-70B \| [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) \| </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> \| Category \| Benchmark (Metric) \| Claude-3.5-Sonnet-1022 \| GPT-4o 0513 \| DeepSeek V3 \| OpenAI o1-mini \| OpenAI o1-1217 \| DeepSeek R1 \| \|----------\|-------------------\|----------------------\|------------\|--------------\|----------------\|------------\|--------------\| \| \| Architecture \| - \| - \| MoE \| - \| - \| MoE \| \| \| # Activated Params \| - \| - \| 37B \| - \| - \| 37B \| \| \| # Total Params \| - \| - \| 671B \| - \| - \| 671B \| \| English \| MMLU (Pass@1) \| 88.3 \| 87.2 \| 88.5 \| 85.2 \| 91.8 \| 90.8 \| \| \| MMLU-Redux (EM) \| 88.9 \| 88.0 \| 89.1 \| 86.7 \| - \| 92.9 \| \| \| MMLU-Pro (EM) \| 78.0 \| 72.6 \| 75.9 \| 80.3 \| - \| 84.0 \| \| \| DROP (3-shot F1) \| 88.3 \| 83.7 \| 91.6 \| 83.9 \| 90.2 \| 92.2 \| \| \| IF-Eval (Prompt Strict) \| 86.5 \| 84.3 \| 86.1 \| 84.8 \| - \| 83.3 \| \| \| GPQA-Diamond (Pass@1) \| 65.0 \| 49.9 \| 59.1 \| 60.0 \| 75.7 \| 71.5 \| \| \| SimpleQA (Correct) \| 28.4 \| 38.2 \| 24.9 \| 7.0 \| 47.0 \| 30.1 \| \| \| FRAMES (Acc.) \| 72.5 \| 80.5 \| 73.3 \| 76.9 \| - \| 82.5 \| \| \| AlpacaEval2.0 (LC-winrate) \| 52.0 \| 51.1 \| 70.0 \| 57.8 \| - \| 87.6 \| \| \| ArenaHard (GPT-4-1106) \| 85.2 \| 80.4 \| 85.5 \| 92.0 \| - \| 92.3 \| \| Code \| LiveCodeBench (Pass@1-COT) \| 33.8 \| 34.2 \| - \| 53.8 \| 63.4 \| 65.9 \| \| \| Codeforces (Percentile) \| 20.3 \| 23.6 \| 58.7 \| 93.4 \| 96.6 \| 96.3 \| \| \| Codeforces (Rating) \| 717 \| 759 \| 1134 \| 1820 \| 2061 \| 2029 \| \| \| SWE Verified (Resolved) \| 50.8 \| 38.8 \| 42.0 \| 41.6 \| 48.9 \| 49.2 \| \| \| Aider-Polyglot (Acc.) \| 45.3 \| 16.0 \| 49.6 \| 32.9 \| 61.7 \| 53.3 \| \| Math \| AIME 2024 (Pass@1) \| 16.0 \| 9.3 \| 39.2 \| 63.6 \| 79.2 \| 79.8 \| \| \| MATH-500 (Pass@1) \| 78.3 \| 74.6 \| 90.2 \| 90.0 \| 96.4 \| 97.3 \| \| \| CNMO 2024 (Pass@1) \| 13.1 \| 10.8 \| 43.2 \| 67.6 \| - \| 78.8 \| \| Chinese \| CLUEWSC (EM) \| 85.4 \| 87.9 \| 90.9 \| 89.9 \| - \| 92.8 \| \| \| C-Eval (EM) \| 76.7 \| 76.0 \| 86.5 \| 68.9 \| - \| 91.8 \| \| \| C-SimpleQA (Correct) \| 55.4 \| 58.7 \| 68.0 \| 40.3 \| - \| 63.7 \| </div> ### Distilled Model Evaluation <div align="center"> \| Model \| AIME 2024 pass@1 \| AIME 2024 cons@64 \| MATH-500 pass@1 \| GPQA Diamond pass@1 \| LiveCodeBench pass@1 \| CodeForces rating \| \|------------------------------------------\|------------------\|-------------------\|-----------------\|----------------------\|----------------------\|-------------------\| \| GPT-4o-0513 \| 9.3 \| 13.4 \| 74.6 \| 49.9 \| 32.9 \| 759 \| \| Claude-3.5-Sonnet-1022 \| 16.0 \| 26.7 \| 78.3 \| 65.0 \| 38.9 \| 717 \| \| o1-mini \| 63.6 \| 80.0 \| 90.0 \| 60.0 \| 53.8 \| 1820 \| \| QwQ-32B-Preview \| 44.0 \| 60.0 \| 90.6 \| 54.5 \| 41.9 \| 1316 \| \| DeepSeek-R1-Distill-Qwen-1.5B \| 28.9 \| 52.7 \| 83.9 \| 33.8 \| 16.9 \| 954 \| \| DeepSeek-R1-Distill-Qwen-7B \| 55.5 \| 83.3 \| 92.8 \| 49.1 \| 37.6 \| 1189 \| \| DeepSeek-R1-Distill-Qwen-14B \| 69.7 \| 80.0 \| 93.9 \| 59.1 \| 53.1 \| 1481 \| \| DeepSeek-R1-Distill-Qwen-32B \| 72.6 \| 83.3 \| 94.3 \| 62.1 \| 57.2 \| 1691 \| \| DeepSeek-R1-Distill-Llama-8B \| 50.4 \| 80.0 \| 89.1 \| 49.0 \| 39.6 \| 1205 \| \| DeepSeek-R1-Distill-Llama-70B \| 70.0 \| 86.7 \| 94.5 \| 65.2 \| 57.5 \| 1633 \| </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output. ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).
oldiday/625b4c61-4c53-4c35-8346-7973e6e5d4d4	oldiday	2025-01-20T23:22:12Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2.5-Coder-7B-Instruct", "base_model:adapter:unsloth/Qwen2.5-Coder-7B-Instruct", "license:apache-2.0", "region:us" ]	null	2025-01-20T22:55:44Z	--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2.5-Coder-7B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 625b4c61-4c53-4c35-8346-7973e6e5d4d4 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2.5-Coder-7B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 5da4fdb4f9d40cf6_train_data.json ds_type: json format: custom path: /workspace/input_data/5da4fdb4f9d40cf6_train_data.json type: field_input: topic; field_instruction: message_1 field_output: message_2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: oldiday/625b4c61-4c53-4c35-8346-7973e6e5d4d4 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 5.0e-05 load_in_4bit: false load_in_8bit: false local_rank: 0 logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 100 micro_batch_size: 8 mlflow_experiment_name: /tmp/5da4fdb4f9d40cf6_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: techspear-hub wandb_mode: online wandb_name: 971c59eb-5de8-4a78-8d22-6a7da4c9ee82 wandb_project: Gradients-On-Six wandb_run: your_name wandb_runid: 971c59eb-5de8-4a78-8d22-6a7da4c9ee82 warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # 625b4c61-4c53-4c35-8346-7973e6e5d4d4 This model is a fine-tuned version of [unsloth/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.5959 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 100 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0017 \| 1 \| 0.6948 \| \| 0.6968 \| 0.0153 \| 9 \| 0.6832 \| \| 0.6562 \| 0.0306 \| 18 \| 0.6471 \| \| 0.6206 \| 0.0459 \| 27 \| 0.6231 \| \| 0.6178 \| 0.0612 \| 36 \| 0.6112 \| \| 0.625 \| 0.0765 \| 45 \| 0.6054 \| \| 0.605 \| 0.0918 \| 54 \| 0.6015 \| \| 0.5935 \| 0.1071 \| 63 \| 0.5990 \| \| 0.6049 \| 0.1224 \| 72 \| 0.5973 \| \| 0.5989 \| 0.1378 \| 81 \| 0.5964 \| \| 0.5937 \| 0.1531 \| 90 \| 0.5960 \| \| 0.6079 \| 0.1684 \| 99 \| 0.5959 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
MayBashendy/ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k18_task5_organization	MayBashendy	2025-01-20T23:22:05Z	5	0	transformers	[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-classification	2025-01-20T23:10:20Z	--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k18_task5_organization results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k18_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2288 - Qwk: 0.0 - Mse: 1.2288 - Rmse: 1.1085 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Qwk \| Mse \| Rmse \| \|:-------------:\|:-------:\|:----:\|:---------------:\|:-------:\|:------:\|:------:\| \| No log \| 0.0465 \| 2 \| 4.0158 \| 0.0034 \| 4.0158 \| 2.0039 \| \| No log \| 0.0930 \| 4 \| 2.2748 \| 0.0271 \| 2.2748 \| 1.5083 \| \| No log \| 0.1395 \| 6 \| 1.6688 \| 0.0329 \| 1.6688 \| 1.2918 \| \| No log \| 0.1860 \| 8 \| 1.2346 \| 0.0967 \| 1.2346 \| 1.1111 \| \| No log \| 0.2326 \| 10 \| 1.1925 \| 0.0731 \| 1.1925 \| 1.0920 \| \| No log \| 0.2791 \| 12 \| 1.1109 \| 0.2150 \| 1.1109 \| 1.0540 \| \| No log \| 0.3256 \| 14 \| 1.0932 \| 0.1884 \| 1.0932 \| 1.0456 \| \| No log \| 0.3721 \| 16 \| 1.2148 \| 0.0380 \| 1.2148 \| 1.1022 \| \| No log \| 0.4186 \| 18 \| 1.2750 \| 0.0 \| 1.2750 \| 1.1292 \| \| No log \| 0.4651 \| 20 \| 1.1496 \| 0.0760 \| 1.1496 \| 1.0722 \| \| No log \| 0.5116 \| 22 \| 1.1357 \| 0.1379 \| 1.1357 \| 1.0657 \| \| No log \| 0.5581 \| 24 \| 1.1803 \| 0.1333 \| 1.1803 \| 1.0864 \| \| No log \| 0.6047 \| 26 \| 1.1167 \| 0.1981 \| 1.1167 \| 1.0568 \| \| No log \| 0.6512 \| 28 \| 1.1559 \| 0.0445 \| 1.1559 \| 1.0751 \| \| No log \| 0.6977 \| 30 \| 1.2187 \| 0.0445 \| 1.2187 \| 1.1039 \| \| No log \| 0.7442 \| 32 \| 1.2232 \| 0.0445 \| 1.2232 \| 1.1060 \| \| No log \| 0.7907 \| 34 \| 1.1709 \| 0.1643 \| 1.1709 \| 1.0821 \| \| No log \| 0.8372 \| 36 \| 1.1800 \| 0.0792 \| 1.1800 \| 1.0863 \| \| No log \| 0.8837 \| 38 \| 1.3732 \| -0.0296 \| 1.3732 \| 1.1719 \| \| No log \| 0.9302 \| 40 \| 1.6117 \| 0.0143 \| 1.6117 \| 1.2695 \| \| No log \| 0.9767 \| 42 \| 1.7119 \| 0.0143 \| 1.7119 \| 1.3084 \| \| No log \| 1.0233 \| 44 \| 1.5533 \| 0.0 \| 1.5533 \| 1.2463 \| \| No log \| 1.0698 \| 46 \| 1.4371 \| -0.0148 \| 1.4371 \| 1.1988 \| \| No log \| 1.1163 \| 48 \| 1.1990 \| 0.0701 \| 1.1990 \| 1.0950 \| \| No log \| 1.1628 \| 50 \| 1.0921 \| 0.2239 \| 1.0921 \| 1.0450 \| \| No log \| 1.2093 \| 52 \| 1.2033 \| 0.0102 \| 1.2033 \| 1.0969 \| \| No log \| 1.2558 \| 54 \| 1.2896 \| -0.1560 \| 1.2896 \| 1.1356 \| \| No log \| 1.3023 \| 56 \| 1.4626 \| -0.0326 \| 1.4626 \| 1.2094 \| \| No log \| 1.3488 \| 58 \| 1.3772 \| -0.0022 \| 1.3772 \| 1.1735 \| \| No log \| 1.3953 \| 60 \| 1.2187 \| 0.1361 \| 1.2187 \| 1.1040 \| \| No log \| 1.4419 \| 62 \| 1.2426 \| 0.0374 \| 1.2426 \| 1.1147 \| \| No log \| 1.4884 \| 64 \| 1.2838 \| 0.0999 \| 1.2838 \| 1.1330 \| \| No log \| 1.5349 \| 66 \| 1.2161 \| 0.0700 \| 1.2161 \| 1.1028 \| \| No log \| 1.5814 \| 68 \| 1.1278 \| 0.2692 \| 1.1278 \| 1.0620 \| \| No log \| 1.6279 \| 70 \| 1.1283 \| 0.1725 \| 1.1283 \| 1.0622 \| \| No log \| 1.6744 \| 72 \| 1.0934 \| 0.2366 \| 1.0934 \| 1.0456 \| \| No log \| 1.7209 \| 74 \| 1.0742 \| 0.1783 \| 1.0742 \| 1.0364 \| \| No log \| 1.7674 \| 76 \| 1.1562 \| 0.1389 \| 1.1562 \| 1.0752 \| \| No log \| 1.8140 \| 78 \| 1.4371 \| -0.0641 \| 1.4371 \| 1.1988 \| \| No log \| 1.8605 \| 80 \| 1.4812 \| -0.0641 \| 1.4812 \| 1.2171 \| \| No log \| 1.9070 \| 82 \| 1.4368 \| -0.0641 \| 1.4368 \| 1.1987 \| \| No log \| 1.9535 \| 84 \| 1.2691 \| -0.0091 \| 1.2691 \| 1.1266 \| \| No log \| 2.0 \| 86 \| 1.2183 \| 0.0542 \| 1.2183 \| 1.1038 \| \| No log \| 2.0465 \| 88 \| 1.2777 \| 0.0188 \| 1.2777 \| 1.1304 \| \| No log \| 2.0930 \| 90 \| 1.4962 \| -0.1067 \| 1.4962 \| 1.2232 \| \| No log \| 2.1395 \| 92 \| 1.5524 \| -0.0747 \| 1.5524 \| 1.2460 \| \| No log \| 2.1860 \| 94 \| 1.3420 \| 0.0587 \| 1.3420 \| 1.1585 \| \| No log \| 2.2326 \| 96 \| 1.1766 \| 0.0164 \| 1.1766 \| 1.0847 \| \| No log \| 2.2791 \| 98 \| 1.1748 \| 0.0164 \| 1.1748 \| 1.0839 \| \| No log \| 2.3256 \| 100 \| 1.3960 \| 0.0464 \| 1.3960 \| 1.1815 \| \| No log \| 2.3721 \| 102 \| 1.7947 \| -0.1843 \| 1.7947 \| 1.3397 \| \| No log \| 2.4186 \| 104 \| 1.9849 \| -0.3093 \| 1.9849 \| 1.4089 \| \| No log \| 2.4651 \| 106 \| 1.9655 \| -0.1443 \| 1.9655 \| 1.4020 \| \| No log \| 2.5116 \| 108 \| 1.7718 \| -0.0788 \| 1.7718 \| 1.3311 \| \| No log \| 2.5581 \| 110 \| 1.5684 \| -0.0167 \| 1.5684 \| 1.2524 \| \| No log \| 2.6047 \| 112 \| 1.6681 \| -0.0688 \| 1.6681 \| 1.2916 \| \| No log \| 2.6512 \| 114 \| 1.8768 \| -0.1179 \| 1.8768 \| 1.3700 \| \| No log \| 2.6977 \| 116 \| 2.0265 \| -0.1122 \| 2.0265 \| 1.4236 \| \| No log \| 2.7442 \| 118 \| 2.0327 \| -0.1111 \| 2.0327 \| 1.4257 \| \| No log \| 2.7907 \| 120 \| 2.0567 \| -0.1154 \| 2.0567 \| 1.4341 \| \| No log \| 2.8372 \| 122 \| 1.8254 \| -0.0655 \| 1.8254 \| 1.3511 \| \| No log \| 2.8837 \| 124 \| 1.5249 \| -0.0735 \| 1.5249 \| 1.2349 \| \| No log \| 2.9302 \| 126 \| 1.7057 \| -0.1078 \| 1.7057 \| 1.3060 \| \| No log \| 2.9767 \| 128 \| 2.0800 \| -0.2156 \| 2.0800 \| 1.4422 \| \| No log \| 3.0233 \| 130 \| 2.3606 \| -0.2661 \| 2.3606 \| 1.5364 \| \| No log \| 3.0698 \| 132 \| 2.2816 \| -0.1571 \| 2.2816 \| 1.5105 \| \| No log \| 3.1163 \| 134 \| 2.0213 \| -0.1658 \| 2.0213 \| 1.4217 \| \| No log \| 3.1628 \| 136 \| 1.9379 \| -0.1254 \| 1.9379 \| 1.3921 \| \| No log \| 3.2093 \| 138 \| 1.8792 \| -0.0892 \| 1.8792 \| 1.3708 \| \| No log \| 3.2558 \| 140 \| 1.8788 \| -0.0849 \| 1.8788 \| 1.3707 \| \| No log \| 3.3023 \| 142 \| 1.7651 \| -0.0762 \| 1.7651 \| 1.3286 \| \| No log \| 3.3488 \| 144 \| 1.5874 \| 0.0294 \| 1.5874 \| 1.2599 \| \| No log \| 3.3953 \| 146 \| 1.5857 \| 0.0294 \| 1.5857 \| 1.2593 \| \| No log \| 3.4419 \| 148 \| 1.6411 \| 0.0279 \| 1.6411 \| 1.2810 \| \| No log \| 3.4884 \| 150 \| 1.7297 \| -0.0267 \| 1.7297 \| 1.3152 \| \| No log \| 3.5349 \| 152 \| 1.7574 \| -0.0757 \| 1.7574 \| 1.3257 \| \| No log \| 3.5814 \| 154 \| 1.7628 \| -0.0397 \| 1.7628 \| 1.3277 \| \| No log \| 3.6279 \| 156 \| 1.7589 \| -0.0508 \| 1.7589 \| 1.3262 \| \| No log \| 3.6744 \| 158 \| 1.7224 \| -0.0378 \| 1.7224 \| 1.3124 \| \| No log \| 3.7209 \| 160 \| 1.7568 \| -0.0514 \| 1.7568 \| 1.3254 \| \| No log \| 3.7674 \| 162 \| 1.8652 \| 0.0726 \| 1.8652 \| 1.3657 \| \| No log \| 3.8140 \| 164 \| 1.9175 \| 0.0402 \| 1.9175 \| 1.3847 \| \| No log \| 3.8605 \| 166 \| 1.8288 \| 0.0400 \| 1.8288 \| 1.3523 \| \| No log \| 3.9070 \| 168 \| 1.8386 \| 0.0562 \| 1.8386 \| 1.3559 \| \| No log \| 3.9535 \| 170 \| 1.8316 \| 0.1342 \| 1.8316 \| 1.3534 \| \| No log \| 4.0 \| 172 \| 1.6490 \| 0.2105 \| 1.6490 \| 1.2842 \| \| No log \| 4.0465 \| 174 \| 1.5040 \| 0.1282 \| 1.5040 \| 1.2264 \| \| No log \| 4.0930 \| 176 \| 1.4472 \| 0.1198 \| 1.4472 \| 1.2030 \| \| No log \| 4.1395 \| 178 \| 1.4452 \| 0.0270 \| 1.4452 \| 1.2022 \| \| No log \| 4.1860 \| 180 \| 1.5266 \| 0.0809 \| 1.5266 \| 1.2356 \| \| No log \| 4.2326 \| 182 \| 1.7969 \| 0.1525 \| 1.7969 \| 1.3405 \| \| No log \| 4.2791 \| 184 \| 1.8888 \| 0.2098 \| 1.8888 \| 1.3744 \| \| No log \| 4.3256 \| 186 \| 1.7524 \| 0.1663 \| 1.7524 \| 1.3238 \| \| No log \| 4.3721 \| 188 \| 1.5118 \| 0.2342 \| 1.5118 \| 1.2296 \| \| No log \| 4.4186 \| 190 \| 1.4063 \| 0.0946 \| 1.4063 \| 1.1859 \| \| No log \| 4.4651 \| 192 \| 1.4720 \| 0.1911 \| 1.4720 \| 1.2133 \| \| No log \| 4.5116 \| 194 \| 1.4943 \| 0.1423 \| 1.4943 \| 1.2224 \| \| No log \| 4.5581 \| 196 \| 1.4577 \| 0.1423 \| 1.4577 \| 1.2073 \| \| No log \| 4.6047 \| 198 \| 1.3969 \| 0.0602 \| 1.3969 \| 1.1819 \| \| No log \| 4.6512 \| 200 \| 1.3910 \| 0.1110 \| 1.3910 \| 1.1794 \| \| No log \| 4.6977 \| 202 \| 1.4709 \| 0.1110 \| 1.4709 \| 1.2128 \| \| No log \| 4.7442 \| 204 \| 1.5518 \| 0.1423 \| 1.5518 \| 1.2457 \| \| No log \| 4.7907 \| 206 \| 1.5587 \| 0.1027 \| 1.5587 \| 1.2485 \| \| No log \| 4.8372 \| 208 \| 1.5567 \| 0.1027 \| 1.5567 \| 1.2477 \| \| No log \| 4.8837 \| 210 \| 1.5407 \| 0.1423 \| 1.5407 \| 1.2413 \| \| No log \| 4.9302 \| 212 \| 1.4965 \| 0.1904 \| 1.4965 \| 1.2233 \| \| No log \| 4.9767 \| 214 \| 1.3997 \| 0.1943 \| 1.3997 \| 1.1831 \| \| No log \| 5.0233 \| 216 \| 1.3739 \| 0.2126 \| 1.3739 \| 1.1721 \| \| No log \| 5.0698 \| 218 \| 1.4697 \| 0.2292 \| 1.4697 \| 1.2123 \| \| No log \| 5.1163 \| 220 \| 1.4313 \| 0.2015 \| 1.4313 \| 1.1964 \| \| No log \| 5.1628 \| 222 \| 1.4067 \| 0.2203 \| 1.4067 \| 1.1860 \| \| No log \| 5.2093 \| 224 \| 1.4171 \| 0.1886 \| 1.4171 \| 1.1904 \| \| No log \| 5.2558 \| 226 \| 1.5312 \| 0.2117 \| 1.5312 \| 1.2374 \| \| No log \| 5.3023 \| 228 \| 1.7132 \| 0.2058 \| 1.7132 \| 1.3089 \| \| No log \| 5.3488 \| 230 \| 1.7432 \| 0.2206 \| 1.7432 \| 1.3203 \| \| No log \| 5.3953 \| 232 \| 1.7312 \| 0.1963 \| 1.7312 \| 1.3158 \| \| No log \| 5.4419 \| 234 \| 1.7061 \| 0.2389 \| 1.7061 \| 1.3062 \| \| No log \| 5.4884 \| 236 \| 1.5999 \| 0.1461 \| 1.5999 \| 1.2649 \| \| No log \| 5.5349 \| 238 \| 1.5428 \| 0.1911 \| 1.5428 \| 1.2421 \| \| No log \| 5.5814 \| 240 \| 1.3937 \| 0.1142 \| 1.3937 \| 1.1806 \| \| No log \| 5.6279 \| 242 \| 1.3014 \| 0.1052 \| 1.3014 \| 1.1408 \| \| No log \| 5.6744 \| 244 \| 1.2485 \| 0.1052 \| 1.2485 \| 1.1174 \| \| No log \| 5.7209 \| 246 \| 1.2400 \| 0.0401 \| 1.2400 \| 1.1136 \| \| No log \| 5.7674 \| 248 \| 1.2161 \| 0.0401 \| 1.2161 \| 1.1028 \| \| No log \| 5.8140 \| 250 \| 1.1670 \| 0.0556 \| 1.1670 \| 1.0803 \| \| No log \| 5.8605 \| 252 \| 1.1944 \| 0.0155 \| 1.1944 \| 1.0929 \| \| No log \| 5.9070 \| 254 \| 1.3146 \| 0.0781 \| 1.3146 \| 1.1466 \| \| No log \| 5.9535 \| 256 \| 1.5336 \| 0.1880 \| 1.5336 \| 1.2384 \| \| No log \| 6.0 \| 258 \| 1.6128 \| 0.2465 \| 1.6128 \| 1.2700 \| \| No log \| 6.0465 \| 260 \| 1.5854 \| 0.2465 \| 1.5854 \| 1.2591 \| \| No log \| 6.0930 \| 262 \| 1.5079 \| 0.1966 \| 1.5079 \| 1.2280 \| \| No log \| 6.1395 \| 264 \| 1.4465 \| 0.1886 \| 1.4465 \| 1.2027 \| \| No log \| 6.1860 \| 266 \| 1.4232 \| 0.2027 \| 1.4232 \| 1.1930 \| \| No log \| 6.2326 \| 268 \| 1.3952 \| 0.2027 \| 1.3952 \| 1.1812 \| \| No log \| 6.2791 \| 270 \| 1.3799 \| 0.2027 \| 1.3799 \| 1.1747 \| \| No log \| 6.3256 \| 272 \| 1.3777 \| 0.2089 \| 1.3777 \| 1.1738 \| \| No log \| 6.3721 \| 274 \| 1.4345 \| 0.2203 \| 1.4345 \| 1.1977 \| \| No log \| 6.4186 \| 276 \| 1.4906 \| 0.2555 \| 1.4906 \| 1.2209 \| \| No log \| 6.4651 \| 278 \| 1.6231 \| 0.2771 \| 1.6231 \| 1.2740 \| \| No log \| 6.5116 \| 280 \| 1.6548 \| 0.2270 \| 1.6548 \| 1.2864 \| \| No log \| 6.5581 \| 282 \| 1.4900 \| 0.1058 \| 1.4900 \| 1.2207 \| \| No log \| 6.6047 \| 284 \| 1.4021 \| 0.0781 \| 1.4021 \| 1.1841 \| \| No log \| 6.6512 \| 286 \| 1.4033 \| 0.0781 \| 1.4033 \| 1.1846 \| \| No log \| 6.6977 \| 288 \| 1.4786 \| 0.2315 \| 1.4786 \| 1.2160 \| \| No log \| 6.7442 \| 290 \| 1.4991 \| 0.2915 \| 1.4991 \| 1.2244 \| \| No log \| 6.7907 \| 292 \| 1.4009 \| 0.2455 \| 1.4009 \| 1.1836 \| \| No log \| 6.8372 \| 294 \| 1.3681 \| 0.2455 \| 1.3681 \| 1.1697 \| \| No log \| 6.8837 \| 296 \| 1.3522 \| 0.2455 \| 1.3522 \| 1.1628 \| \| No log \| 6.9302 \| 298 \| 1.3574 \| 0.2506 \| 1.3574 \| 1.1651 \| \| No log \| 6.9767 \| 300 \| 1.4005 \| 0.2506 \| 1.4005 \| 1.1834 \| \| No log \| 7.0233 \| 302 \| 1.4022 \| 0.2203 \| 1.4022 \| 1.1841 \| \| No log \| 7.0698 \| 304 \| 1.3696 \| 0.1886 \| 1.3696 \| 1.1703 \| \| No log \| 7.1163 \| 306 \| 1.3437 \| 0.1886 \| 1.3437 \| 1.1592 \| \| No log \| 7.1628 \| 308 \| 1.3611 \| 0.1886 \| 1.3611 \| 1.1667 \| \| No log \| 7.2093 \| 310 \| 1.4336 \| 0.2260 \| 1.4336 \| 1.1973 \| \| No log \| 7.2558 \| 312 \| 1.4199 \| 0.1886 \| 1.4199 \| 1.1916 \| \| No log \| 7.3023 \| 314 \| 1.3951 \| 0.1886 \| 1.3951 \| 1.1811 \| \| No log \| 7.3488 \| 316 \| 1.3180 \| 0.1316 \| 1.3180 \| 1.1481 \| \| No log \| 7.3953 \| 318 \| 1.3180 \| 0.1548 \| 1.3180 \| 1.1480 \| \| No log \| 7.4419 \| 320 \| 1.4264 \| 0.1980 \| 1.4264 \| 1.1943 \| \| No log \| 7.4884 \| 322 \| 1.7047 \| 0.2315 \| 1.7047 \| 1.3057 \| \| No log \| 7.5349 \| 324 \| 1.7725 \| 0.2116 \| 1.7725 \| 1.3313 \| \| No log \| 7.5814 \| 326 \| 1.6272 \| 0.2363 \| 1.6272 \| 1.2756 \| \| No log \| 7.6279 \| 328 \| 1.5210 \| 0.2395 \| 1.5210 \| 1.2333 \| \| No log \| 7.6744 \| 330 \| 1.4116 \| 0.2027 \| 1.4116 \| 1.1881 \| \| No log \| 7.7209 \| 332 \| 1.3019 \| 0.1622 \| 1.3019 \| 1.1410 \| \| No log \| 7.7674 \| 334 \| 1.2722 \| 0.1622 \| 1.2722 \| 1.1279 \| \| No log \| 7.8140 \| 336 \| 1.3118 \| 0.1473 \| 1.3118 \| 1.1453 \| \| No log \| 7.8605 \| 338 \| 1.3846 \| 0.1886 \| 1.3846 \| 1.1767 \| \| No log \| 7.9070 \| 340 \| 1.3764 \| 0.1886 \| 1.3764 \| 1.1732 \| \| No log \| 7.9535 \| 342 \| 1.3974 \| 0.1886 \| 1.3974 \| 1.1821 \| \| No log \| 8.0 \| 344 \| 1.4388 \| 0.2506 \| 1.4388 \| 1.1995 \| \| No log \| 8.0465 \| 346 \| 1.4310 \| 0.2126 \| 1.4310 \| 1.1962 \| \| No log \| 8.0930 \| 348 \| 1.4688 \| 0.2424 \| 1.4688 \| 1.2119 \| \| No log \| 8.1395 \| 350 \| 1.4365 \| 0.2424 \| 1.4365 \| 1.1986 \| \| No log \| 8.1860 \| 352 \| 1.3914 \| 0.1814 \| 1.3914 \| 1.1796 \| \| No log \| 8.2326 \| 354 \| 1.3864 \| 0.1814 \| 1.3864 \| 1.1775 \| \| No log \| 8.2791 \| 356 \| 1.3512 \| 0.1552 \| 1.3512 \| 1.1624 \| \| No log \| 8.3256 \| 358 \| 1.3229 \| 0.1473 \| 1.3229 \| 1.1502 \| \| No log \| 8.3721 \| 360 \| 1.3194 \| 0.1552 \| 1.3194 \| 1.1486 \| \| No log \| 8.4186 \| 362 \| 1.3151 \| 0.1552 \| 1.3151 \| 1.1468 \| \| No log \| 8.4651 \| 364 \| 1.4077 \| 0.2640 \| 1.4077 \| 1.1865 \| \| No log \| 8.5116 \| 366 \| 1.5515 \| 0.2223 \| 1.5515 \| 1.2456 \| \| No log \| 8.5581 \| 368 \| 1.6437 \| 0.1978 \| 1.6437 \| 1.2821 \| \| No log \| 8.6047 \| 370 \| 1.7347 \| 0.2414 \| 1.7347 \| 1.3171 \| \| No log \| 8.6512 \| 372 \| 1.8221 \| 0.2224 \| 1.8221 \| 1.3499 \| \| No log \| 8.6977 \| 374 \| 1.6633 \| 0.2481 \| 1.6633 \| 1.2897 \| \| No log \| 8.7442 \| 376 \| 1.4077 \| 0.1898 \| 1.4077 \| 1.1865 \| \| No log \| 8.7907 \| 378 \| 1.3061 \| 0.0931 \| 1.3061 \| 1.1428 \| \| No log \| 8.8372 \| 380 \| 1.2222 \| 0.0160 \| 1.2222 \| 1.1055 \| \| No log \| 8.8837 \| 382 \| 1.2155 \| 0.0160 \| 1.2155 \| 1.1025 \| \| No log \| 8.9302 \| 384 \| 1.2623 \| 0.0445 \| 1.2623 \| 1.1235 \| \| No log \| 8.9767 \| 386 \| 1.3885 \| 0.1552 \| 1.3885 \| 1.1783 \| \| No log \| 9.0233 \| 388 \| 1.4957 \| 0.2795 \| 1.4957 \| 1.2230 \| \| No log \| 9.0698 \| 390 \| 1.3858 \| 0.2640 \| 1.3858 \| 1.1772 \| \| No log \| 9.1163 \| 392 \| 1.2752 \| 0.1961 \| 1.2752 \| 1.1293 \| \| No log \| 9.1628 \| 394 \| 1.2226 \| 0.1961 \| 1.2226 \| 1.1057 \| \| No log \| 9.2093 \| 396 \| 1.1909 \| 0.1697 \| 1.1909 \| 1.0913 \| \| No log \| 9.2558 \| 398 \| 1.1800 \| 0.1552 \| 1.1800 \| 1.0863 \| \| No log \| 9.3023 \| 400 \| 1.1418 \| 0.1202 \| 1.1418 \| 1.0686 \| \| No log \| 9.3488 \| 402 \| 1.1663 \| 0.1552 \| 1.1663 \| 1.0800 \| \| No log \| 9.3953 \| 404 \| 1.1681 \| 0.1552 \| 1.1681 \| 1.0808 \| \| No log \| 9.4419 \| 406 \| 1.2341 \| 0.1552 \| 1.2341 \| 1.1109 \| \| No log \| 9.4884 \| 408 \| 1.3670 \| 0.2647 \| 1.3670 \| 1.1692 \| \| No log \| 9.5349 \| 410 \| 1.4327 \| 0.2465 \| 1.4327 \| 1.1970 \| \| No log \| 9.5814 \| 412 \| 1.4405 \| 0.2465 \| 1.4405 \| 1.2002 \| \| No log \| 9.6279 \| 414 \| 1.4134 \| 0.2602 \| 1.4134 \| 1.1889 \| \| No log \| 9.6744 \| 416 \| 1.3585 \| 0.2506 \| 1.3585 \| 1.1656 \| \| No log \| 9.7209 \| 418 \| 1.3508 \| 0.1886 \| 1.3508 \| 1.1623 \| \| No log \| 9.7674 \| 420 \| 1.3591 \| 0.1952 \| 1.3591 \| 1.1658 \| \| No log \| 9.8140 \| 422 \| 1.3131 \| 0.1952 \| 1.3131 \| 1.1459 \| \| No log \| 9.8605 \| 424 \| 1.3215 \| 0.1886 \| 1.3215 \| 1.1496 \| \| No log \| 9.9070 \| 426 \| 1.4130 \| 0.2506 \| 1.4130 \| 1.1887 \| \| No log \| 9.9535 \| 428 \| 1.3922 \| 0.1310 \| 1.3922 \| 1.1799 \| \| No log \| 10.0 \| 430 \| 1.4431 \| 0.1428 \| 1.4431 \| 1.2013 \| \| No log \| 10.0465 \| 432 \| 1.5592 \| 0.2117 \| 1.5592 \| 1.2487 \| \| No log \| 10.0930 \| 434 \| 1.5122 \| 0.2292 \| 1.5122 \| 1.2297 \| \| No log \| 10.1395 \| 436 \| 1.4766 \| 0.2647 \| 1.4766 \| 1.2151 \| \| No log \| 10.1860 \| 438 \| 1.3397 \| 0.2315 \| 1.3397 \| 1.1574 \| \| No log \| 10.2326 \| 440 \| 1.2819 \| 0.1886 \| 1.2819 \| 1.1322 \| \| No log \| 10.2791 \| 442 \| 1.2954 \| 0.1700 \| 1.2954 \| 1.1381 \| \| No log \| 10.3256 \| 444 \| 1.3657 \| 0.2315 \| 1.3657 \| 1.1686 \| \| No log \| 10.3721 \| 446 \| 1.4421 \| 0.2315 \| 1.4421 \| 1.2009 \| \| No log \| 10.4186 \| 448 \| 1.4492 \| 0.2015 \| 1.4492 \| 1.2038 \| \| No log \| 10.4651 \| 450 \| 1.3858 \| 0.1202 \| 1.3858 \| 1.1772 \| \| No log \| 10.5116 \| 452 \| 1.3957 \| 0.0833 \| 1.3957 \| 1.1814 \| \| No log \| 10.5581 \| 454 \| 1.4324 \| 0.0833 \| 1.4324 \| 1.1968 \| \| No log \| 10.6047 \| 456 \| 1.5332 \| 0.1552 \| 1.5332 \| 1.2382 \| \| No log \| 10.6512 \| 458 \| 1.5961 \| 0.1952 \| 1.5961 \| 1.2634 \| \| No log \| 10.6977 \| 460 \| 1.6255 \| 0.1595 \| 1.6255 \| 1.2750 \| \| No log \| 10.7442 \| 462 \| 1.5482 \| 0.1634 \| 1.5482 \| 1.2443 \| \| No log \| 10.7907 \| 464 \| 1.4990 \| 0.1407 \| 1.4990 \| 1.2243 \| \| No log \| 10.8372 \| 466 \| 1.4643 \| 0.1473 \| 1.4643 \| 1.2101 \| \| No log \| 10.8837 \| 468 \| 1.4725 \| 0.1407 \| 1.4725 \| 1.2135 \| \| No log \| 10.9302 \| 470 \| 1.4109 \| 0.0781 \| 1.4109 \| 1.1878 \| \| No log \| 10.9767 \| 472 \| 1.3841 \| 0.1142 \| 1.3841 \| 1.1765 \| \| No log \| 11.0233 \| 474 \| 1.4789 \| 0.2239 \| 1.4789 \| 1.2161 \| \| No log \| 11.0698 \| 476 \| 1.5078 \| 0.2126 \| 1.5078 \| 1.2279 \| \| No log \| 11.1163 \| 478 \| 1.5220 \| 0.1486 \| 1.5220 \| 1.2337 \| \| No log \| 11.1628 \| 480 \| 1.4774 \| 0.1552 \| 1.4774 \| 1.2155 \| \| No log \| 11.2093 \| 482 \| 1.4310 \| 0.1486 \| 1.4310 \| 1.1962 \| \| No log \| 11.2558 \| 484 \| 1.4349 \| 0.2424 \| 1.4349 \| 1.1979 \| \| No log \| 11.3023 \| 486 \| 1.3930 \| 0.2424 \| 1.3930 \| 1.1803 \| \| No log \| 11.3488 \| 488 \| 1.3830 \| 0.2126 \| 1.3830 \| 1.1760 \| \| No log \| 11.3953 \| 490 \| 1.3882 \| 0.1407 \| 1.3882 \| 1.1782 \| \| No log \| 11.4419 \| 492 \| 1.3933 \| 0.1473 \| 1.3933 \| 1.1804 \| \| No log \| 11.4884 \| 494 \| 1.3616 \| 0.1473 \| 1.3616 \| 1.1669 \| \| No log \| 11.5349 \| 496 \| 1.3209 \| 0.1473 \| 1.3209 \| 1.1493 \| \| No log \| 11.5814 \| 498 \| 1.2735 \| 0.0401 \| 1.2735 \| 1.1285 \| \| 0.3529 \| 11.6279 \| 500 \| 1.2606 \| 0.0401 \| 1.2606 \| 1.1228 \| \| 0.3529 \| 11.6744 \| 502 \| 1.2457 \| 0.0401 \| 1.2457 \| 1.1161 \| \| 0.3529 \| 11.7209 \| 504 \| 1.2800 \| 0.0401 \| 1.2800 \| 1.1314 \| \| 0.3529 \| 11.7674 \| 506 \| 1.3705 \| 0.1052 \| 1.3705 \| 1.1707 \| \| 0.3529 \| 11.8140 \| 508 \| 1.4245 \| 0.1351 \| 1.4245 \| 1.1935 \| \| 0.3529 \| 11.8605 \| 510 \| 1.4236 \| 0.1351 \| 1.4236 \| 1.1932 \| \| 0.3529 \| 11.9070 \| 512 \| 1.4686 \| 0.1142 \| 1.4686 \| 1.2118 \| \| 0.3529 \| 11.9535 \| 514 \| 1.5206 \| 0.2126 \| 1.5206 \| 1.2331 \| \| 0.3529 \| 12.0 \| 516 \| 1.4949 \| 0.2126 \| 1.4949 \| 1.2226 \| \| 0.3529 \| 12.0465 \| 518 \| 1.4517 \| 0.1562 \| 1.4517 \| 1.2048 \| \| 0.3529 \| 12.0930 \| 520 \| 1.4807 \| 0.2126 \| 1.4807 \| 1.2169 \| \| 0.3529 \| 12.1395 \| 522 \| 1.5059 \| 0.2239 \| 1.5059 \| 1.2272 \| \| 0.3529 \| 12.1860 \| 524 \| 1.6043 \| 0.2391 \| 1.6043 \| 1.2666 \| \| 0.3529 \| 12.2326 \| 526 \| 1.5695 \| 0.2391 \| 1.5695 \| 1.2528 \| \| 0.3529 \| 12.2791 \| 528 \| 1.4159 \| 0.2424 \| 1.4159 \| 1.1899 \| \| 0.3529 \| 12.3256 \| 530 \| 1.3219 \| 0.1486 \| 1.3219 \| 1.1497 \| \| 0.3529 \| 12.3721 \| 532 \| 1.3401 \| 0.2065 \| 1.3401 \| 1.1576 \| \| 0.3529 \| 12.4186 \| 534 \| 1.3882 \| 0.2424 \| 1.3882 \| 1.1782 \| \| 0.3529 \| 12.4651 \| 536 \| 1.3911 \| 0.2126 \| 1.3911 \| 1.1795 \| \| 0.3529 \| 12.5116 \| 538 \| 1.3736 \| 0.2506 \| 1.3736 \| 1.1720 \| \| 0.3529 \| 12.5581 \| 540 \| 1.3963 \| 0.2424 \| 1.3963 \| 1.1816 \| \| 0.3529 \| 12.6047 \| 542 \| 1.4377 \| 0.2709 \| 1.4377 \| 1.1990 \| \| 0.3529 \| 12.6512 \| 544 \| 1.3851 \| 0.2126 \| 1.3851 \| 1.1769 \| \| 0.3529 \| 12.6977 \| 546 \| 1.3814 \| 0.2126 \| 1.3814 \| 1.1753 \| \| 0.3529 \| 12.7442 \| 548 \| 1.4161 \| 0.2424 \| 1.4161 \| 1.1900 \| \| 0.3529 \| 12.7907 \| 550 \| 1.4834 \| 0.2690 \| 1.4834 \| 1.2180 \| \| 0.3529 \| 12.8372 \| 552 \| 1.5133 \| 0.2731 \| 1.5133 \| 1.2302 \| \| 0.3529 \| 12.8837 \| 554 \| 1.4658 \| 0.2690 \| 1.4658 \| 1.2107 \| \| 0.3529 \| 12.9302 \| 556 \| 1.3921 \| 0.2315 \| 1.3921 \| 1.1799 \| \| 0.3529 \| 12.9767 \| 558 \| 1.3701 \| 0.2506 \| 1.3701 \| 1.1705 \| \| 0.3529 \| 13.0233 \| 560 \| 1.4052 \| 0.2795 \| 1.4052 \| 1.1854 \| \| 0.3529 \| 13.0698 \| 562 \| 1.4930 \| 0.2602 \| 1.4930 \| 1.2219 \| \| 0.3529 \| 13.1163 \| 564 \| 1.5950 \| 0.2731 \| 1.5950 \| 1.2629 \| \| 0.3529 \| 13.1628 \| 566 \| 1.6690 \| 0.3232 \| 1.6690 \| 1.2919 \| \| 0.3529 \| 13.2093 \| 568 \| 1.6193 \| 0.3172 \| 1.6193 \| 1.2725 \| \| 0.3529 \| 13.2558 \| 570 \| 1.5363 \| 0.2752 \| 1.5363 \| 1.2395 \| \| 0.3529 \| 13.3023 \| 572 \| 1.4229 \| 0.0833 \| 1.4229 \| 1.1928 \| \| 0.3529 \| 13.3488 \| 574 \| 1.3506 \| 0.0833 \| 1.3506 \| 1.1622 \| \| 0.3529 \| 13.3953 \| 576 \| 1.3492 \| 0.0833 \| 1.3492 \| 1.1616 \| \| 0.3529 \| 13.4419 \| 578 \| 1.3896 \| 0.0833 \| 1.3896 \| 1.1788 \| \| 0.3529 \| 13.4884 \| 580 \| 1.4283 \| 0.0781 \| 1.4283 \| 1.1951 \| \| 0.3529 \| 13.5349 \| 582 \| 1.4724 \| 0.1142 \| 1.4724 \| 1.2134 \| \| 0.3529 \| 13.5814 \| 584 \| 1.4633 \| 0.1142 \| 1.4633 \| 1.2097 \| \| 0.3529 \| 13.6279 \| 586 \| 1.4614 \| 0.1142 \| 1.4614 \| 1.2089 \| \| 0.3529 \| 13.6744 \| 588 \| 1.4658 \| 0.1486 \| 1.4658 \| 1.2107 \| \| 0.3529 \| 13.7209 \| 590 \| 1.4604 \| 0.1228 \| 1.4604 \| 1.2085 \| \| 0.3529 \| 13.7674 \| 592 \| 1.3979 \| 0.1228 \| 1.3979 \| 1.1823 \| \| 0.3529 \| 13.8140 \| 594 \| 1.3791 \| 0.1228 \| 1.3791 \| 1.1744 \| \| 0.3529 \| 13.8605 \| 596 \| 1.4188 \| 0.2126 \| 1.4188 \| 1.1911 \| \| 0.3529 \| 13.9070 \| 598 \| 1.4513 \| 0.2709 \| 1.4513 \| 1.2047 \| \| 0.3529 \| 13.9535 \| 600 \| 1.4620 \| 0.2709 \| 1.4620 \| 1.2092 \| \| 0.3529 \| 14.0 \| 602 \| 1.3948 \| 0.2424 \| 1.3948 \| 1.1810 \| \| 0.3529 \| 14.0465 \| 604 \| 1.3774 \| 0.2126 \| 1.3774 \| 1.1736 \| \| 0.3529 \| 14.0930 \| 606 \| 1.3843 \| 0.2424 \| 1.3843 \| 1.1765 \| \| 0.3529 \| 14.1395 \| 608 \| 1.3777 \| 0.1814 \| 1.3777 \| 1.1738 \| \| 0.3529 \| 14.1860 \| 610 \| 1.3095 \| 0.0781 \| 1.3095 \| 1.1443 \| \| 0.3529 \| 14.2326 \| 612 \| 1.2472 \| 0.0401 \| 1.2472 \| 1.1168 \| \| 0.3529 \| 14.2791 \| 614 \| 1.2122 \| 0.0401 \| 1.2122 \| 1.1010 \| \| 0.3529 \| 14.3256 \| 616 \| 1.1979 \| 0.0 \| 1.1979 \| 1.0945 \| \| 0.3529 \| 14.3721 \| 618 \| 1.2415 \| 0.0401 \| 1.2415 \| 1.1142 \| \| 0.3529 \| 14.4186 \| 620 \| 1.2917 \| 0.0781 \| 1.2917 \| 1.1365 \| \| 0.3529 \| 14.4651 \| 622 \| 1.3478 \| 0.1310 \| 1.3478 \| 1.1609 \| \| 0.3529 \| 14.5116 \| 624 \| 1.4031 \| 0.1634 \| 1.4031 \| 1.1845 \| \| 0.3529 \| 14.5581 \| 626 \| 1.4545 \| 0.2522 \| 1.4545 \| 1.2060 \| \| 0.3529 \| 14.6047 \| 628 \| 1.4968 \| 0.2522 \| 1.4968 \| 1.2234 \| \| 0.3529 \| 14.6512 \| 630 \| 1.5350 \| 0.2522 \| 1.5350 \| 1.2389 \| \| 0.3529 \| 14.6977 \| 632 \| 1.5023 \| 0.2239 \| 1.5023 \| 1.2257 \| \| 0.3529 \| 14.7442 \| 634 \| 1.3888 \| 0.0401 \| 1.3888 \| 1.1785 \| \| 0.3529 \| 14.7907 \| 636 \| 1.2818 \| 0.0 \| 1.2818 \| 1.1322 \| \| 0.3529 \| 14.8372 \| 638 \| 1.2042 \| 0.0 \| 1.2042 \| 1.0974 \| \| 0.3529 \| 14.8837 \| 640 \| 1.1860 \| 0.0 \| 1.1860 \| 1.0890 \| \| 0.3529 \| 14.9302 \| 642 \| 1.2288 \| 0.0 \| 1.2288 \| 1.1085 \| ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1
datlaaaaaaa/e8cccbe1-c136-409a-a725-9f09127c1a3f	datlaaaaaaa	2025-01-20T23:21:58Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:54:13Z	--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: e8cccbe1-c136-409a-a725-9f09127c1a3f results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 854bca96bed40197_train_data.json ds_type: json format: custom path: /workspace/input_data/854bca96bed40197_train_data.json type: field_input: state_before field_instruction: tactic field_output: state_after format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: datlaaaaaaa/e8cccbe1-c136-409a-a725-9f09127c1a3f hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/854bca96bed40197_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: cff9d1c5-a847-4707-b347-d0451baf6b24 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: cff9d1c5-a847-4707-b347-d0451baf6b24 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # e8cccbe1-c136-409a-a725-9f09127c1a3f This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.3004 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0849 \| 0.0077 \| 200 \| 0.3004 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
adamo1139/DeepSeek-R1-Distill-Qwen-1.5B-6bpw-exl2	adamo1139	2025-01-20T23:21:56Z	8	0	null	[ "qwen2", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "6-bit", "exl2", "region:us" ]	null	2025-01-20T22:51:16Z	--- base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --- # DeepSeek-R1 <!-- markdownlint-disable first-line-h1 --> <!-- markdownlint-disable html --> <!-- markdownlint-disable no-duplicate-header --> <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- Post-Training: Large-Scale Reinforcement Learning on the Base Model - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- Distillation: Smaller Models Can Be Powerful Too - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> \| Model \| #Total Params \| #Activated Params \| Context Length \| Download \| \| :------------: \| :------------: \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Zero \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) \| \| DeepSeek-R1 \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) \| </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> \| Model \| Base Model \| Download \| \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Distill-Qwen-1.5B \| [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) \| \| DeepSeek-R1-Distill-Qwen-7B \| [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) \| \| DeepSeek-R1-Distill-Llama-8B \| [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) \| \| DeepSeek-R1-Distill-Qwen-14B \| [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) \| \|DeepSeek-R1-Distill-Qwen-32B \| [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) \| \| DeepSeek-R1-Distill-Llama-70B \| [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) \| </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> \| Category \| Benchmark (Metric) \| Claude-3.5-Sonnet-1022 \| GPT-4o 0513 \| DeepSeek V3 \| OpenAI o1-mini \| OpenAI o1-1217 \| DeepSeek R1 \| \|----------\|-------------------\|----------------------\|------------\|--------------\|----------------\|------------\|--------------\| \| \| Architecture \| - \| - \| MoE \| - \| - \| MoE \| \| \| # Activated Params \| - \| - \| 37B \| - \| - \| 37B \| \| \| # Total Params \| - \| - \| 671B \| - \| - \| 671B \| \| English \| MMLU (Pass@1) \| 88.3 \| 87.2 \| 88.5 \| 85.2 \| 91.8 \| 90.8 \| \| \| MMLU-Redux (EM) \| 88.9 \| 88.0 \| 89.1 \| 86.7 \| - \| 92.9 \| \| \| MMLU-Pro (EM) \| 78.0 \| 72.6 \| 75.9 \| 80.3 \| - \| 84.0 \| \| \| DROP (3-shot F1) \| 88.3 \| 83.7 \| 91.6 \| 83.9 \| 90.2 \| 92.2 \| \| \| IF-Eval (Prompt Strict) \| 86.5 \| 84.3 \| 86.1 \| 84.8 \| - \| 83.3 \| \| \| GPQA-Diamond (Pass@1) \| 65.0 \| 49.9 \| 59.1 \| 60.0 \| 75.7 \| 71.5 \| \| \| SimpleQA (Correct) \| 28.4 \| 38.2 \| 24.9 \| 7.0 \| 47.0 \| 30.1 \| \| \| FRAMES (Acc.) \| 72.5 \| 80.5 \| 73.3 \| 76.9 \| - \| 82.5 \| \| \| AlpacaEval2.0 (LC-winrate) \| 52.0 \| 51.1 \| 70.0 \| 57.8 \| - \| 87.6 \| \| \| ArenaHard (GPT-4-1106) \| 85.2 \| 80.4 \| 85.5 \| 92.0 \| - \| 92.3 \| \| Code \| LiveCodeBench (Pass@1-COT) \| 33.8 \| 34.2 \| - \| 53.8 \| 63.4 \| 65.9 \| \| \| Codeforces (Percentile) \| 20.3 \| 23.6 \| 58.7 \| 93.4 \| 96.6 \| 96.3 \| \| \| Codeforces (Rating) \| 717 \| 759 \| 1134 \| 1820 \| 2061 \| 2029 \| \| \| SWE Verified (Resolved) \| 50.8 \| 38.8 \| 42.0 \| 41.6 \| 48.9 \| 49.2 \| \| \| Aider-Polyglot (Acc.) \| 45.3 \| 16.0 \| 49.6 \| 32.9 \| 61.7 \| 53.3 \| \| Math \| AIME 2024 (Pass@1) \| 16.0 \| 9.3 \| 39.2 \| 63.6 \| 79.2 \| 79.8 \| \| \| MATH-500 (Pass@1) \| 78.3 \| 74.6 \| 90.2 \| 90.0 \| 96.4 \| 97.3 \| \| \| CNMO 2024 (Pass@1) \| 13.1 \| 10.8 \| 43.2 \| 67.6 \| - \| 78.8 \| \| Chinese \| CLUEWSC (EM) \| 85.4 \| 87.9 \| 90.9 \| 89.9 \| - \| 92.8 \| \| \| C-Eval (EM) \| 76.7 \| 76.0 \| 86.5 \| 68.9 \| - \| 91.8 \| \| \| C-SimpleQA (Correct) \| 55.4 \| 58.7 \| 68.0 \| 40.3 \| - \| 63.7 \| </div> ### Distilled Model Evaluation <div align="center"> \| Model \| AIME 2024 pass@1 \| AIME 2024 cons@64 \| MATH-500 pass@1 \| GPQA Diamond pass@1 \| LiveCodeBench pass@1 \| CodeForces rating \| \|------------------------------------------\|------------------\|-------------------\|-----------------\|----------------------\|----------------------\|-------------------\| \| GPT-4o-0513 \| 9.3 \| 13.4 \| 74.6 \| 49.9 \| 32.9 \| 759 \| \| Claude-3.5-Sonnet-1022 \| 16.0 \| 26.7 \| 78.3 \| 65.0 \| 38.9 \| 717 \| \| o1-mini \| 63.6 \| 80.0 \| 90.0 \| 60.0 \| 53.8 \| 1820 \| \| QwQ-32B-Preview \| 44.0 \| 60.0 \| 90.6 \| 54.5 \| 41.9 \| 1316 \| \| DeepSeek-R1-Distill-Qwen-1.5B \| 28.9 \| 52.7 \| 83.9 \| 33.8 \| 16.9 \| 954 \| \| DeepSeek-R1-Distill-Qwen-7B \| 55.5 \| 83.3 \| 92.8 \| 49.1 \| 37.6 \| 1189 \| \| DeepSeek-R1-Distill-Qwen-14B \| 69.7 \| 80.0 \| 93.9 \| 59.1 \| 53.1 \| 1481 \| \| DeepSeek-R1-Distill-Qwen-32B \| 72.6 \| 83.3 \| 94.3 \| 62.1 \| 57.2 \| 1691 \| \| DeepSeek-R1-Distill-Llama-8B \| 50.4 \| 80.0 \| 89.1 \| 49.0 \| 39.6 \| 1205 \| \| DeepSeek-R1-Distill-Llama-70B \| 70.0 \| 86.7 \| 94.5 \| 65.2 \| 57.5 \| 1633 \| </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output. ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).
adamo1139/DeepSeek-R1-Distill-Qwen-1.5B-8bpw-exl2	adamo1139	2025-01-20T23:21:21Z	27	0	null	[ "qwen2", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "8-bit", "exl2", "region:us" ]	null	2025-01-20T22:50:28Z	--- base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --- # DeepSeek-R1 <!-- markdownlint-disable first-line-h1 --> <!-- markdownlint-disable html --> <!-- markdownlint-disable no-duplicate-header --> <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- Post-Training: Large-Scale Reinforcement Learning on the Base Model - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- Distillation: Smaller Models Can Be Powerful Too - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> \| Model \| #Total Params \| #Activated Params \| Context Length \| Download \| \| :------------: \| :------------: \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Zero \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) \| \| DeepSeek-R1 \| 671B \| 37B \| 128K \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) \| </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> \| Model \| Base Model \| Download \| \| :------------: \| :------------: \| :------------: \| \| DeepSeek-R1-Distill-Qwen-1.5B \| [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) \| \| DeepSeek-R1-Distill-Qwen-7B \| [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) \| \| DeepSeek-R1-Distill-Llama-8B \| [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) \| \| DeepSeek-R1-Distill-Qwen-14B \| [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) \| \|DeepSeek-R1-Distill-Qwen-32B \| [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) \| \| DeepSeek-R1-Distill-Llama-70B \| [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) \| [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) \| </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> \| Category \| Benchmark (Metric) \| Claude-3.5-Sonnet-1022 \| GPT-4o 0513 \| DeepSeek V3 \| OpenAI o1-mini \| OpenAI o1-1217 \| DeepSeek R1 \| \|----------\|-------------------\|----------------------\|------------\|--------------\|----------------\|------------\|--------------\| \| \| Architecture \| - \| - \| MoE \| - \| - \| MoE \| \| \| # Activated Params \| - \| - \| 37B \| - \| - \| 37B \| \| \| # Total Params \| - \| - \| 671B \| - \| - \| 671B \| \| English \| MMLU (Pass@1) \| 88.3 \| 87.2 \| 88.5 \| 85.2 \| 91.8 \| 90.8 \| \| \| MMLU-Redux (EM) \| 88.9 \| 88.0 \| 89.1 \| 86.7 \| - \| 92.9 \| \| \| MMLU-Pro (EM) \| 78.0 \| 72.6 \| 75.9 \| 80.3 \| - \| 84.0 \| \| \| DROP (3-shot F1) \| 88.3 \| 83.7 \| 91.6 \| 83.9 \| 90.2 \| 92.2 \| \| \| IF-Eval (Prompt Strict) \| 86.5 \| 84.3 \| 86.1 \| 84.8 \| - \| 83.3 \| \| \| GPQA-Diamond (Pass@1) \| 65.0 \| 49.9 \| 59.1 \| 60.0 \| 75.7 \| 71.5 \| \| \| SimpleQA (Correct) \| 28.4 \| 38.2 \| 24.9 \| 7.0 \| 47.0 \| 30.1 \| \| \| FRAMES (Acc.) \| 72.5 \| 80.5 \| 73.3 \| 76.9 \| - \| 82.5 \| \| \| AlpacaEval2.0 (LC-winrate) \| 52.0 \| 51.1 \| 70.0 \| 57.8 \| - \| 87.6 \| \| \| ArenaHard (GPT-4-1106) \| 85.2 \| 80.4 \| 85.5 \| 92.0 \| - \| 92.3 \| \| Code \| LiveCodeBench (Pass@1-COT) \| 33.8 \| 34.2 \| - \| 53.8 \| 63.4 \| 65.9 \| \| \| Codeforces (Percentile) \| 20.3 \| 23.6 \| 58.7 \| 93.4 \| 96.6 \| 96.3 \| \| \| Codeforces (Rating) \| 717 \| 759 \| 1134 \| 1820 \| 2061 \| 2029 \| \| \| SWE Verified (Resolved) \| 50.8 \| 38.8 \| 42.0 \| 41.6 \| 48.9 \| 49.2 \| \| \| Aider-Polyglot (Acc.) \| 45.3 \| 16.0 \| 49.6 \| 32.9 \| 61.7 \| 53.3 \| \| Math \| AIME 2024 (Pass@1) \| 16.0 \| 9.3 \| 39.2 \| 63.6 \| 79.2 \| 79.8 \| \| \| MATH-500 (Pass@1) \| 78.3 \| 74.6 \| 90.2 \| 90.0 \| 96.4 \| 97.3 \| \| \| CNMO 2024 (Pass@1) \| 13.1 \| 10.8 \| 43.2 \| 67.6 \| - \| 78.8 \| \| Chinese \| CLUEWSC (EM) \| 85.4 \| 87.9 \| 90.9 \| 89.9 \| - \| 92.8 \| \| \| C-Eval (EM) \| 76.7 \| 76.0 \| 86.5 \| 68.9 \| - \| 91.8 \| \| \| C-SimpleQA (Correct) \| 55.4 \| 58.7 \| 68.0 \| 40.3 \| - \| 63.7 \| </div> ### Distilled Model Evaluation <div align="center"> \| Model \| AIME 2024 pass@1 \| AIME 2024 cons@64 \| MATH-500 pass@1 \| GPQA Diamond pass@1 \| LiveCodeBench pass@1 \| CodeForces rating \| \|------------------------------------------\|------------------\|-------------------\|-----------------\|----------------------\|----------------------\|-------------------\| \| GPT-4o-0513 \| 9.3 \| 13.4 \| 74.6 \| 49.9 \| 32.9 \| 759 \| \| Claude-3.5-Sonnet-1022 \| 16.0 \| 26.7 \| 78.3 \| 65.0 \| 38.9 \| 717 \| \| o1-mini \| 63.6 \| 80.0 \| 90.0 \| 60.0 \| 53.8 \| 1820 \| \| QwQ-32B-Preview \| 44.0 \| 60.0 \| 90.6 \| 54.5 \| 41.9 \| 1316 \| \| DeepSeek-R1-Distill-Qwen-1.5B \| 28.9 \| 52.7 \| 83.9 \| 33.8 \| 16.9 \| 954 \| \| DeepSeek-R1-Distill-Qwen-7B \| 55.5 \| 83.3 \| 92.8 \| 49.1 \| 37.6 \| 1189 \| \| DeepSeek-R1-Distill-Qwen-14B \| 69.7 \| 80.0 \| 93.9 \| 59.1 \| 53.1 \| 1481 \| \| DeepSeek-R1-Distill-Qwen-32B \| 72.6 \| 83.3 \| 94.3 \| 62.1 \| 57.2 \| 1691 \| \| DeepSeek-R1-Distill-Llama-8B \| 50.4 \| 80.0 \| 89.1 \| 49.0 \| 39.6 \| 1205 \| \| DeepSeek-R1-Distill-Llama-70B \| 70.0 \| 86.7 \| 94.5 \| 65.2 \| 57.5 \| 1633 \| </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output. ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).
LHRuig/cinestyle	LHRuig	2025-01-20T23:20:52Z	5	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:20:41Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # cinestyle <Gallery /> ## Model description cinestyle lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/cinestyle/tree/main) them in the Files & versions tab.
LHRuig/cinedrama	LHRuig	2025-01-20T23:20:46Z	8	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T23:19:20Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # cinedrama <Gallery /> ## Model description cinedrama lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/cinedrama/tree/main) them in the Files & versions tab.
kaizen9/phi-1_5_HQ_3000_20k	kaizen9	2025-01-20T23:20:32Z	15	0	transformers	[ "transformers", "safetensors", "phi", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-01-20T23:16:36Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
lesso08/7bc9e54e-f755-40b7-a740-c391d742641d	lesso08	2025-01-20T23:20:20Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2-7B-Instruct", "base_model:adapter:unsloth/Qwen2-7B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:22:58Z	--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2-7B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 7bc9e54e-f755-40b7-a740-c391d742641d results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2-7B-Instruct bf16: true chat_template: llama3 datasets: - data_files: - 6e60a538f672529c_train_data.json ds_type: json format: custom path: /workspace/input_data/6e60a538f672529c_train_data.json type: field_input: communityName field_instruction: label field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso08/7bc9e54e-f755-40b7-a740-c391d742641d hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/6e60a538f672529c_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 3eb92360-4e77-4cc9-9ffa-0e03d7ea7423 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 3eb92360-4e77-4cc9-9ffa-0e03d7ea7423 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 7bc9e54e-f755-40b7-a740-c391d742641d This model is a fine-tuned version of [unsloth/Qwen2-7B-Instruct](https://huggingface.co/unsloth/Qwen2-7B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.0001 \| 1 \| nan \| \| 0.0 \| 0.0004 \| 5 \| nan \| \| 0.0 \| 0.0008 \| 10 \| nan \| \| 0.0 \| 0.0012 \| 15 \| nan \| \| 0.0 \| 0.0016 \| 20 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
mrHungddddh/9b1deaac-4565-48f2-a511-89a1ab96e3e3	mrHungddddh	2025-01-20T23:19:50Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:04:02Z	--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 9b1deaac-4565-48f2-a511-89a1ab96e3e3 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - a7412d8c8f805ddf_train_data.json ds_type: json format: custom path: /workspace/input_data/a7412d8c8f805ddf_train_data.json type: field_instruction: premise field_output: hypothesis format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: mrHungddddh/9b1deaac-4565-48f2-a511-89a1ab96e3e3 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/a7412d8c8f805ddf_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ba28db81-f399-44e5-bdef-7af8dcf5a4ca warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 9b1deaac-4565-48f2-a511-89a1ab96e3e3 This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2632 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.7427 \| 0.0304 \| 200 \| 1.2632 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
rawsh/q1-3B-PRIME	rawsh	2025-01-20T23:19:31Z	523	1	null	[ "safetensors", "qwen2", "text-generation", "conversational", "en", "dataset:PRIME-RL/Eurus-2-RL-Data", "base_model:PowerInfer/SmallThinker-3B-Preview", "base_model:finetune:PowerInfer/SmallThinker-3B-Preview", "region:us" ]	text-generation	2025-01-16T03:26:54Z	--- base_model: - Qwen/Qwen2.5-3B-Instruct - PowerInfer/SmallThinker-3B-Preview datasets: - PRIME-RL/Eurus-2-RL-Data language: - en pipeline_tag: text-generation --- # q1-3B-PRIME q1-3B-PRIME, a small reasoning model trained with reinforcement learning. Trained using SmallThinker-3B-Preview as a base model (Qwen2.5-3B-Instruct full finetuned on QwQ reasoning traces) for a roughly ~22.5% improvement on the test set in 120 training steps. (Note: lots of performance left on the table since PRIME saturates after 300+ steps.) # Benchmark Performance ## Math \| Model \| AIME24 \| AMC23 \| MATH-500 \| \|---------\|--------\|-------\|-------\| \| Qwen2.5-3B-Instruct \| 6.67 \| 45 \| - \| \| SmallThinker-3B-Preview\| 16.667 \| 57.5 \| - \| \| q1-3B-PRIME \| 26.667 \| 67.5 \| 64.8 \| \| Eurus-7B-PRIME \| 26.667 \| 57.8 \| 79.2 \| \| GPT-4o \| 9.3 \| 45.8 \| 76.4 \| ## Coding \| Model \| HumanEval \| Leetcode \| \|---------\|--------\|-------\| \| Qwen2.5-3B-Instruct \| 74.4 \| - \| \| q1-3B-PRIME \| 71.95 \| 20.55 \| \| GPT-4o \| 90.2 \| - \|
laquythang/efe15045-778f-4dfa-8d03-17ab758f41b0	laquythang	2025-01-20T23:18:35Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:04:13Z	--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: efe15045-778f-4dfa-8d03-17ab758f41b0 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - a7412d8c8f805ddf_train_data.json ds_type: json format: custom path: /workspace/input_data/a7412d8c8f805ddf_train_data.json type: field_instruction: premise field_output: hypothesis format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: laquythang/efe15045-778f-4dfa-8d03-17ab758f41b0 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/a7412d8c8f805ddf_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ba28db81-f399-44e5-bdef-7af8dcf5a4ca warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # efe15045-778f-4dfa-8d03-17ab758f41b0 This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2649 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.7354 \| 0.0304 \| 200 \| 1.2649 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
nadejdatarabukina/7623d9ae-b2cb-4a91-801e-e6ab04be6251	nadejdatarabukina	2025-01-20T23:18:27Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:unsloth/Meta-Llama-3.1-8B", "base_model:adapter:unsloth/Meta-Llama-3.1-8B", "license:llama3.1", "region:us" ]	null	2025-01-20T22:59:19Z	--- library_name: peft license: llama3.1 base_model: unsloth/Meta-Llama-3.1-8B tags: - axolotl - generated_from_trainer model-index: - name: 7623d9ae-b2cb-4a91-801e-e6ab04be6251 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Meta-Llama-3.1-8B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 91e193d3dca1611f_train_data.json ds_type: json format: custom path: /workspace/input_data/91e193d3dca1611f_train_data.json type: field_input: parent_id field_instruction: role field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: null eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: nadejdatarabukina/7623d9ae-b2cb-4a91-801e-e6ab04be6251 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 75GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/91e193d3dca1611f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 856a9aac-189f-40f7-b27c-c5616995b0d1 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 856a9aac-189f-40f7-b27c-c5616995b0d1 warmup_steps: 10 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 7623d9ae-b2cb-4a91-801e-e6ab04be6251 This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0004 \| 1 \| nan \| \| 0.0 \| 0.0018 \| 5 \| nan \| \| 0.0 \| 0.0035 \| 10 \| nan \| \| 0.0 \| 0.0053 \| 15 \| nan \| \| 0.0 \| 0.0071 \| 20 \| nan \| \| 0.0 \| 0.0088 \| 25 \| nan \| \| 0.0 \| 0.0106 \| 30 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
John6666/real-dream-sdxlpony14-sdxl	John6666	2025-01-20T23:18:26Z	659	0	diffusers	[ "diffusers", "safetensors", "text-to-image", "stable-diffusion", "stable-diffusion-xl", "realistic", "photorealistic", "photo", "pony", "en", "base_model:luisrguerra/real-dream-xl-pony-releases", "base_model:finetune:luisrguerra/real-dream-xl-pony-releases", "license:other", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionXLPipeline", "region:us" ]	text-to-image	2025-01-20T23:13:33Z	--- license: other license_name: faipl-1.0-sd license_link: https://freedevproject.org/faipl-1.0-sd/ language: - en library_name: diffusers pipeline_tag: text-to-image tags: - text-to-image - stable-diffusion - stable-diffusion-xl - realistic - photorealistic - photo - pony base_model: luisrguerra/real-dream-xl-pony-releases --- Original model is [here](https://civitai.com/models/153568/real-dream?modelVersionId=1308507). The author is [here](https://huggingface.co/luisrguerra). This model created by [sinatra](https://civitai.com/user/sinatra).
nblinh/94055e20-ee90-4d03-9da3-41f4e7349a5a	nblinh	2025-01-20T23:17:56Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4", "base_model:adapter:MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T21:52:01Z	--- library_name: peft base_model: MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4 tags: - axolotl - generated_from_trainer model-index: - name: 94055e20-ee90-4d03-9da3-41f4e7349a5a results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 1983306ea4f53c9d_train_data.json ds_type: json format: custom path: /workspace/input_data/1983306ea4f53c9d_train_data.json type: field_input: prompt field_instruction: instruction field_output: response format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nblinh/94055e20-ee90-4d03-9da3-41f4e7349a5a hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/1983306ea4f53c9d_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 75177f7d-3059-4092-918d-8e9c49bae6b5 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 75177f7d-3059-4092-918d-8e9c49bae6b5 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 94055e20-ee90-4d03-9da3-41f4e7349a5a This model is a fine-tuned version of [MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4](https://huggingface.co/MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.3520 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.333 \| 0.0334 \| 200 \| 0.3520 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
Monday-Someday/mobilenet_v2_1.0_224-finetuned-ISIC-dec2024test	Monday-Someday	2025-01-20T23:17:03Z	8	0	transformers	[ "transformers", "tensorboard", "safetensors", "mobilenet_v2", "image-classification", "generated_from_trainer", "dataset:imagefolder", "base_model:google/mobilenet_v2_1.0_224", "base_model:finetune:google/mobilenet_v2_1.0_224", "license:other", "model-index", "autotrain_compatible", "endpoints_compatible", "region:us" ]	image-classification	2025-01-20T02:12:01Z	--- library_name: transformers license: other base_model: google/mobilenet_v2_1.0_224 tags: - generated_from_trainer datasets: - imagefolder metrics: - accuracy model-index: - name: mobilenet_v2_1.0_224-finetuned-ISIC-dec2024test results: - task: name: Image Classification type: image-classification dataset: name: imagefolder type: imagefolder config: default split: train args: default metrics: - name: Accuracy type: accuracy value: 0.9276220745449292 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # mobilenet_v2_1.0_224-finetuned-ISIC-dec2024test This model is a fine-tuned version of [google/mobilenet_v2_1.0_224](https://huggingface.co/google/mobilenet_v2_1.0_224) on the imagefolder dataset. It achieves the following results on the evaluation set: - Loss: 0.1763 - Accuracy: 0.9276 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 32 - eval_batch_size: 32 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 128 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1 - num_epochs: 5 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| \|:-------------:\|:------:\|:----:\|:---------------:\|:--------:\| \| 0.9055 \| 0.9985 \| 486 \| 0.1955 \| 0.9195 \| \| 0.8797 \| 1.9985 \| 972 \| 0.2074 \| 0.9138 \| \| 0.8144 \| 2.9985 \| 1458 \| 0.1797 \| 0.9263 \| \| 0.9243 \| 3.9985 \| 1944 \| 0.1862 \| 0.9233 \| \| 0.8199 \| 4.9985 \| 2430 \| 0.1763 \| 0.9276 \| ### Framework versions - Transformers 4.47.1 - Pytorch 2.5.1+cpu - Datasets 3.2.0 - Tokenizers 0.21.0
lesso15/858b3fe8-837d-4d0b-908d-af0eb85b1273	lesso15	2025-01-20T23:15:30Z	8	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/CodeLlama-13b-hf-flash", "base_model:adapter:NousResearch/CodeLlama-13b-hf-flash", "4-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:43:29Z	--- library_name: peft base_model: NousResearch/CodeLlama-13b-hf-flash tags: - axolotl - generated_from_trainer model-index: - name: 858b3fe8-837d-4d0b-908d-af0eb85b1273 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/CodeLlama-13b-hf-flash bf16: auto chat_template: llama3 datasets: - data_files: - 9ff4e3b24bf3b2a4_train_data.json ds_type: json format: custom path: /workspace/input_data/9ff4e3b24bf3b2a4_train_data.json type: field_input: sentence1 field_instruction: phrase1 field_output: sentence2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: true gradient_checkpointing: false group_by_length: false hub_model_id: lesso15/858b3fe8-837d-4d0b-908d-af0eb85b1273 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: true load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/9ff4e3b24bf3b2a4_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # 858b3fe8-837d-4d0b-908d-af0eb85b1273 This model is a fine-tuned version of [NousResearch/CodeLlama-13b-hf-flash](https://huggingface.co/NousResearch/CodeLlama-13b-hf-flash) on the None dataset. It achieves the following results on the evaluation set: - Loss: 3.0180 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0003 \| 1 \| 4.9637 \| \| 5.0445 \| 0.0014 \| 5 \| 4.8259 \| \| 4.2696 \| 0.0029 \| 10 \| 3.8969 \| \| 3.0239 \| 0.0043 \| 15 \| 3.1104 \| \| 3.3092 \| 0.0057 \| 20 \| 3.0548 \| \| 2.8685 \| 0.0071 \| 25 \| 3.0180 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
kokovova/e5efb00d-850a-4f79-b6b3-19433c4b5d28	kokovova	2025-01-20T23:14:22Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:princeton-nlp/Sheared-LLaMA-1.3B", "base_model:adapter:princeton-nlp/Sheared-LLaMA-1.3B", "license:apache-2.0", "region:us" ]	null	2025-01-20T22:15:38Z	--- library_name: peft license: apache-2.0 base_model: princeton-nlp/Sheared-LLaMA-1.3B tags: - axolotl - generated_from_trainer model-index: - name: e5efb00d-850a-4f79-b6b3-19433c4b5d28 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: princeton-nlp/Sheared-LLaMA-1.3B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 6d346ae45cb7310f_train_data.json ds_type: json format: custom path: /workspace/input_data/6d346ae45cb7310f_train_data.json type: field_input: problem field_instruction: prompt field_output: solution format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: kokovova/e5efb00d-850a-4f79-b6b3-19433c4b5d28 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/6d346ae45cb7310f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 293f4171-e6c9-4854-a803-09018c88d137 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 293f4171-e6c9-4854-a803-09018c88d137 warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # e5efb00d-850a-4f79-b6b3-19433c4b5d28 This model is a fine-tuned version of [princeton-nlp/Sheared-LLaMA-1.3B](https://huggingface.co/princeton-nlp/Sheared-LLaMA-1.3B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0000 \| 1 \| nan \| \| 0.0 \| 0.0001 \| 5 \| nan \| \| 0.0 \| 0.0002 \| 10 \| nan \| \| 0.0 \| 0.0003 \| 15 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
nhungphammmmm/d9757073-d7e1-40b9-be8d-0c20a4559179	nhungphammmmm	2025-01-20T23:13:54Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "base_model:adapter:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:02:39Z	--- library_name: peft base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f tags: - axolotl - generated_from_trainer model-index: - name: d9757073-d7e1-40b9-be8d-0c20a4559179 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json ds_type: json format: custom path: /workspace/input_data/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nhungphammmmm/d9757073-d7e1-40b9-be8d-0c20a4559179 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075526eb-32e0-4485-aab7-014e4d302171 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075526eb-32e0-4485-aab7-014e4d302171 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # d9757073-d7e1-40b9-be8d-0c20a4559179 This model is a fine-tuned version of [samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f](https://huggingface.co/samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.1318 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.2385 \| 0.0407 \| 200 \| 1.1318 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
nhunglaaaaaaa/133de99e-2f53-4b9b-b8d2-edea6793573f	nhunglaaaaaaa	2025-01-20T23:13:15Z	8	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T23:03:55Z	--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 133de99e-2f53-4b9b-b8d2-edea6793573f results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - a7412d8c8f805ddf_train_data.json ds_type: json format: custom path: /workspace/input_data/a7412d8c8f805ddf_train_data.json type: field_instruction: premise field_output: hypothesis format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nhunglaaaaaaa/133de99e-2f53-4b9b-b8d2-edea6793573f hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/a7412d8c8f805ddf_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ba28db81-f399-44e5-bdef-7af8dcf5a4ca warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 133de99e-2f53-4b9b-b8d2-edea6793573f This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2642 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.7557 \| 0.0304 \| 200 \| 1.2642 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
sergioalves/94dd5542-37d0-4a0a-b042-711c0d084791	sergioalves	2025-01-20T23:12:29Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "base_model:adapter:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "region:us" ]	null	2025-01-20T23:02:39Z	--- library_name: peft base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f tags: - axolotl - generated_from_trainer model-index: - name: 94dd5542-37d0-4a0a-b042-711c0d084791 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json ds_type: json format: custom path: /workspace/input_data/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: sergioalves/94dd5542-37d0-4a0a-b042-711c0d084791 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 75GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_hf output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075526eb-32e0-4485-aab7-014e4d302171 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075526eb-32e0-4485-aab7-014e4d302171 warmup_steps: 10 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 94dd5542-37d0-4a0a-b042-711c0d084791 This model is a fine-tuned version of [samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f](https://huggingface.co/samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_HF with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0002 \| 1 \| nan \| \| 0.0 \| 0.0010 \| 5 \| nan \| \| 0.0 \| 0.0020 \| 10 \| nan \| \| 0.0 \| 0.0031 \| 15 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
LHRuig/hyperreal	LHRuig	2025-01-20T23:10:45Z	7	0	diffusers	[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]	text-to-image	2025-01-20T22:56:28Z	--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # hyperreal <Gallery /> ## Model description hyperreal lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/hyperreal/tree/main) them in the Files & versions tab.
MayBashendy/ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k17_task5_organization	MayBashendy	2025-01-20T23:09:52Z	5	0	transformers	[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-classification	2025-01-20T23:00:25Z	--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k17_task5_organization results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k17_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2639 - Qwk: 0.1814 - Mse: 1.2639 - Rmse: 1.1242 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Qwk \| Mse \| Rmse \| \|:-------------:\|:-------:\|:----:\|:---------------:\|:-------:\|:------:\|:------:\| \| No log \| 0.0488 \| 2 \| 3.9152 \| -0.0319 \| 3.9152 \| 1.9787 \| \| No log \| 0.0976 \| 4 \| 2.3161 \| 0.0372 \| 2.3161 \| 1.5219 \| \| No log \| 0.1463 \| 6 \| 2.0454 \| 0.0260 \| 2.0454 \| 1.4302 \| \| No log \| 0.1951 \| 8 \| 1.6301 \| 0.0100 \| 1.6301 \| 1.2768 \| \| No log \| 0.2439 \| 10 \| 1.2168 \| 0.1091 \| 1.2168 \| 1.1031 \| \| No log \| 0.2927 \| 12 \| 1.1692 \| 0.1901 \| 1.1692 \| 1.0813 \| \| No log \| 0.3415 \| 14 \| 1.2690 \| 0.1205 \| 1.2690 \| 1.1265 \| \| No log \| 0.3902 \| 16 \| 1.3094 \| 0.0613 \| 1.3094 \| 1.1443 \| \| No log \| 0.4390 \| 18 \| 1.2486 \| 0.0374 \| 1.2486 \| 1.1174 \| \| No log \| 0.4878 \| 20 \| 1.1393 \| 0.1304 \| 1.1393 \| 1.0674 \| \| No log \| 0.5366 \| 22 \| 1.0717 \| 0.1167 \| 1.0717 \| 1.0352 \| \| No log \| 0.5854 \| 24 \| 1.0785 \| 0.0888 \| 1.0785 \| 1.0385 \| \| No log \| 0.6341 \| 26 \| 1.1062 \| 0.0445 \| 1.1062 \| 1.0518 \| \| No log \| 0.6829 \| 28 \| 1.0674 \| 0.1203 \| 1.0674 \| 1.0331 \| \| No log \| 0.7317 \| 30 \| 0.9890 \| 0.1107 \| 0.9890 \| 0.9945 \| \| No log \| 0.7805 \| 32 \| 0.9718 \| 0.3221 \| 0.9718 \| 0.9858 \| \| No log \| 0.8293 \| 34 \| 1.1094 \| 0.1576 \| 1.1094 \| 1.0533 \| \| No log \| 0.8780 \| 36 \| 1.0090 \| 0.2392 \| 1.0090 \| 1.0045 \| \| No log \| 0.9268 \| 38 \| 0.9661 \| 0.3935 \| 0.9661 \| 0.9829 \| \| No log \| 0.9756 \| 40 \| 0.9632 \| 0.2834 \| 0.9632 \| 0.9814 \| \| No log \| 1.0244 \| 42 \| 1.0046 \| 0.2217 \| 1.0046 \| 1.0023 \| \| No log \| 1.0732 \| 44 \| 1.0990 \| 0.1189 \| 1.0990 \| 1.0483 \| \| No log \| 1.1220 \| 46 \| 1.1539 \| 0.0761 \| 1.1539 \| 1.0742 \| \| No log \| 1.1707 \| 48 \| 1.1020 \| 0.0823 \| 1.1020 \| 1.0498 \| \| No log \| 1.2195 \| 50 \| 1.0902 \| 0.1350 \| 1.0902 \| 1.0441 \| \| No log \| 1.2683 \| 52 \| 1.1294 \| 0.1003 \| 1.1294 \| 1.0627 \| \| No log \| 1.3171 \| 54 \| 1.1120 \| 0.2145 \| 1.1120 \| 1.0545 \| \| No log \| 1.3659 \| 56 \| 1.0491 \| 0.2912 \| 1.0491 \| 1.0243 \| \| No log \| 1.4146 \| 58 \| 0.9670 \| 0.3048 \| 0.9670 \| 0.9834 \| \| No log \| 1.4634 \| 60 \| 0.9772 \| 0.1330 \| 0.9772 \| 0.9885 \| \| No log \| 1.5122 \| 62 \| 1.0123 \| 0.0888 \| 1.0123 \| 1.0061 \| \| No log \| 1.5610 \| 64 \| 1.0275 \| 0.0888 \| 1.0275 \| 1.0136 \| \| No log \| 1.6098 \| 66 \| 1.0126 \| 0.0888 \| 1.0126 \| 1.0063 \| \| No log \| 1.6585 \| 68 \| 0.9315 \| 0.2672 \| 0.9315 \| 0.9651 \| \| No log \| 1.7073 \| 70 \| 0.9216 \| 0.2895 \| 0.9216 \| 0.9600 \| \| No log \| 1.7561 \| 72 \| 1.0374 \| 0.2392 \| 1.0374 \| 1.0186 \| \| No log \| 1.8049 \| 74 \| 1.0876 \| 0.3283 \| 1.0876 \| 1.0429 \| \| No log \| 1.8537 \| 76 \| 1.1482 \| 0.3283 \| 1.1482 \| 1.0716 \| \| No log \| 1.9024 \| 78 \| 1.2706 \| 0.2926 \| 1.2706 \| 1.1272 \| \| No log \| 1.9512 \| 80 \| 1.4393 \| 0.2588 \| 1.4393 \| 1.1997 \| \| No log \| 2.0 \| 82 \| 1.6033 \| 0.2016 \| 1.6033 \| 1.2662 \| \| No log \| 2.0488 \| 84 \| 1.7020 \| 0.2026 \| 1.7020 \| 1.3046 \| \| No log \| 2.0976 \| 86 \| 1.6749 \| 0.1914 \| 1.6749 \| 1.2942 \| \| No log \| 2.1463 \| 88 \| 1.5513 \| 0.0973 \| 1.5513 \| 1.2455 \| \| No log \| 2.1951 \| 90 \| 1.4070 \| 0.2149 \| 1.4070 \| 1.1862 \| \| No log \| 2.2439 \| 92 \| 1.4139 \| 0.1362 \| 1.4139 \| 1.1891 \| \| No log \| 2.2927 \| 94 \| 1.4569 \| 0.1297 \| 1.4569 \| 1.2070 \| \| No log \| 2.3415 \| 96 \| 1.3699 \| 0.0976 \| 1.3699 \| 1.1704 \| \| No log \| 2.3902 \| 98 \| 1.1940 \| 0.1434 \| 1.1940 \| 1.0927 \| \| No log \| 2.4390 \| 100 \| 1.0175 \| 0.2263 \| 1.0175 \| 1.0087 \| \| No log \| 2.4878 \| 102 \| 0.9743 \| 0.1857 \| 0.9743 \| 0.9871 \| \| No log \| 2.5366 \| 104 \| 0.9971 \| 0.1707 \| 0.9971 \| 0.9986 \| \| No log \| 2.5854 \| 106 \| 1.0930 \| 0.1970 \| 1.0930 \| 1.0454 \| \| No log \| 2.6341 \| 108 \| 1.1941 \| 0.0571 \| 1.1941 \| 1.0928 \| \| No log \| 2.6829 \| 110 \| 1.3024 \| -0.0297 \| 1.3024 \| 1.1412 \| \| No log \| 2.7317 \| 112 \| 1.3887 \| -0.0460 \| 1.3887 \| 1.1784 \| \| No log \| 2.7805 \| 114 \| 1.4502 \| 0.1194 \| 1.4502 \| 1.2043 \| \| No log \| 2.8293 \| 116 \| 1.4877 \| 0.2313 \| 1.4877 \| 1.2197 \| \| No log \| 2.8780 \| 118 \| 1.5875 \| 0.2638 \| 1.5875 \| 1.2599 \| \| No log \| 2.9268 \| 120 \| 1.6965 \| 0.2252 \| 1.6965 \| 1.3025 \| \| No log \| 2.9756 \| 122 \| 1.7232 \| 0.1078 \| 1.7232 \| 1.3127 \| \| No log \| 3.0244 \| 124 \| 1.6639 \| 0.1207 \| 1.6639 \| 1.2899 \| \| No log \| 3.0732 \| 126 \| 1.5461 \| 0.2004 \| 1.5461 \| 1.2434 \| \| No log \| 3.1220 \| 128 \| 1.4334 \| 0.1438 \| 1.4334 \| 1.1973 \| \| No log \| 3.1707 \| 130 \| 1.4259 \| 0.0712 \| 1.4259 \| 1.1941 \| \| No log \| 3.2195 \| 132 \| 1.4356 \| -0.0541 \| 1.4356 \| 1.1982 \| \| No log \| 3.2683 \| 134 \| 1.4579 \| -0.0939 \| 1.4579 \| 1.2074 \| \| No log \| 3.3171 \| 136 \| 1.4491 \| 0.0147 \| 1.4491 \| 1.2038 \| \| No log \| 3.3659 \| 138 \| 1.5209 \| 0.2752 \| 1.5209 \| 1.2332 \| \| No log \| 3.4146 \| 140 \| 1.5553 \| 0.2110 \| 1.5553 \| 1.2471 \| \| No log \| 3.4634 \| 142 \| 1.5332 \| 0.2317 \| 1.5332 \| 1.2382 \| \| No log \| 3.5122 \| 144 \| 1.4563 \| 0.2437 \| 1.4563 \| 1.2068 \| \| No log \| 3.5610 \| 146 \| 1.4090 \| 0.2694 \| 1.4090 \| 1.1870 \| \| No log \| 3.6098 \| 148 \| 1.4642 \| 0.2694 \| 1.4642 \| 1.2101 \| \| No log \| 3.6585 \| 150 \| 1.6547 \| 0.2644 \| 1.6547 \| 1.2864 \| \| No log \| 3.7073 \| 152 \| 1.9515 \| 0.2247 \| 1.9515 \| 1.3970 \| \| No log \| 3.7561 \| 154 \| 2.0626 \| 0.1896 \| 2.0626 \| 1.4362 \| \| No log \| 3.8049 \| 156 \| 1.9816 \| 0.2127 \| 1.9816 \| 1.4077 \| \| No log \| 3.8537 \| 158 \| 1.7125 \| 0.2252 \| 1.7125 \| 1.3086 \| \| No log \| 3.9024 \| 160 \| 1.4900 \| 0.2221 \| 1.4900 \| 1.2206 \| \| No log \| 3.9512 \| 162 \| 1.3005 \| 0.2424 \| 1.3005 \| 1.1404 \| \| No log \| 4.0 \| 164 \| 1.3190 \| 0.2709 \| 1.3190 \| 1.1485 \| \| No log \| 4.0488 \| 166 \| 1.3899 \| 0.2752 \| 1.3899 \| 1.1789 \| \| No log \| 4.0976 \| 168 \| 1.5366 \| 0.2363 \| 1.5366 \| 1.2396 \| \| No log \| 4.1463 \| 170 \| 1.6811 \| 0.2448 \| 1.6811 \| 1.2966 \| \| No log \| 4.1951 \| 172 \| 1.6823 \| 0.2406 \| 1.6823 \| 1.2971 \| \| No log \| 4.2439 \| 174 \| 1.6922 \| 0.2406 \| 1.6922 \| 1.3009 \| \| No log \| 4.2927 \| 176 \| 1.6000 \| 0.2869 \| 1.6000 \| 1.2649 \| \| No log \| 4.3415 \| 178 \| 1.5276 \| 0.2869 \| 1.5276 \| 1.2360 \| \| No log \| 4.3902 \| 180 \| 1.5637 \| 0.2869 \| 1.5637 \| 1.2505 \| \| No log \| 4.4390 \| 182 \| 1.5011 \| 0.2869 \| 1.5011 \| 1.2252 \| \| No log \| 4.4878 \| 184 \| 1.3543 \| 0.2982 \| 1.3543 \| 1.1637 \| \| No log \| 4.5366 \| 186 \| 1.1786 \| 0.2709 \| 1.1786 \| 1.0856 \| \| No log \| 4.5854 \| 188 \| 1.1007 \| 0.2795 \| 1.1007 \| 1.0492 \| \| No log \| 4.6341 \| 190 \| 1.1784 \| 0.2709 \| 1.1784 \| 1.0855 \| \| No log \| 4.6829 \| 192 \| 1.3811 \| 0.2982 \| 1.3811 \| 1.1752 \| \| No log \| 4.7317 \| 194 \| 1.5727 \| 0.2832 \| 1.5727 \| 1.2541 \| \| No log \| 4.7805 \| 196 \| 1.6856 \| 0.2270 \| 1.6856 \| 1.2983 \| \| No log \| 4.8293 \| 198 \| 1.6794 \| 0.2437 \| 1.6794 \| 1.2959 \| \| No log \| 4.8780 \| 200 \| 1.5514 \| 0.2522 \| 1.5514 \| 1.2456 \| \| No log \| 4.9268 \| 202 \| 1.4239 \| 0.2709 \| 1.4239 \| 1.1933 \| \| No log \| 4.9756 \| 204 \| 1.2668 \| 0.1142 \| 1.2668 \| 1.1255 \| \| No log \| 5.0244 \| 206 \| 1.2415 \| 0.1142 \| 1.2415 \| 1.1142 \| \| No log \| 5.0732 \| 208 \| 1.3106 \| 0.2065 \| 1.3106 \| 1.1448 \| \| No log \| 5.1220 \| 210 \| 1.4415 \| 0.2424 \| 1.4415 \| 1.2006 \| \| No log \| 5.1707 \| 212 \| 1.6183 \| 0.1832 \| 1.6183 \| 1.2721 \| \| No log \| 5.2195 \| 214 \| 1.7023 \| 0.1533 \| 1.7023 \| 1.3047 \| \| No log \| 5.2683 \| 216 \| 1.6420 \| 0.1142 \| 1.6420 \| 1.2814 \| \| No log \| 5.3171 \| 218 \| 1.5048 \| 0.1562 \| 1.5048 \| 1.2267 \| \| No log \| 5.3659 \| 220 \| 1.4030 \| 0.1744 \| 1.4030 \| 1.1845 \| \| No log \| 5.4146 \| 222 \| 1.3956 \| 0.1744 \| 1.3956 \| 1.1814 \| \| No log \| 5.4634 \| 224 \| 1.4510 \| 0.2065 \| 1.4510 \| 1.2046 \| \| No log \| 5.5122 \| 226 \| 1.5505 \| 0.2522 \| 1.5505 \| 1.2452 \| \| No log \| 5.5610 \| 228 \| 1.5876 \| 0.2568 \| 1.5876 \| 1.2600 \| \| No log \| 5.6098 \| 230 \| 1.4723 \| 0.1880 \| 1.4723 \| 1.2134 \| \| No log \| 5.6585 \| 232 \| 1.4053 \| 0.1562 \| 1.4053 \| 1.1854 \| \| No log \| 5.7073 \| 234 \| 1.4550 \| 0.1943 \| 1.4550 \| 1.2063 \| \| No log \| 5.7561 \| 236 \| 1.5212 \| 0.1943 \| 1.5212 \| 1.2334 \| \| No log \| 5.8049 \| 238 \| 1.5631 \| 0.1634 \| 1.5631 \| 1.2502 \| \| No log \| 5.8537 \| 240 \| 1.6752 \| 0.1141 \| 1.6752 \| 1.2943 \| \| No log \| 5.9024 \| 242 \| 1.6585 \| 0.1634 \| 1.6585 \| 1.2878 \| \| No log \| 5.9512 \| 244 \| 1.5822 \| 0.2004 \| 1.5822 \| 1.2578 \| \| No log \| 6.0 \| 246 \| 1.4101 \| 0.2522 \| 1.4101 \| 1.1875 \| \| No log \| 6.0488 \| 248 \| 1.2845 \| 0.2184 \| 1.2845 \| 1.1333 \| \| No log \| 6.0976 \| 250 \| 1.3726 \| 0.3052 \| 1.3726 \| 1.1716 \| \| No log \| 6.1463 \| 252 \| 1.5895 \| 0.2527 \| 1.5895 \| 1.2608 \| \| No log \| 6.1951 \| 254 \| 1.7285 \| 0.2419 \| 1.7285 \| 1.3147 \| \| No log \| 6.2439 \| 256 \| 1.7297 \| 0.2419 \| 1.7297 \| 1.3152 \| \| No log \| 6.2927 \| 258 \| 1.5794 \| 0.2296 \| 1.5794 \| 1.2567 \| \| No log \| 6.3415 \| 260 \| 1.5724 \| 0.2296 \| 1.5724 \| 1.2539 \| \| No log \| 6.3902 \| 262 \| 1.5028 \| 0.3052 \| 1.5028 \| 1.2259 \| \| No log \| 6.4390 \| 264 \| 1.4208 \| 0.2474 \| 1.4208 \| 1.1920 \| \| No log \| 6.4878 \| 266 \| 1.3507 \| 0.2126 \| 1.3507 \| 1.1622 \| \| No log \| 6.5366 \| 268 \| 1.3940 \| 0.2424 \| 1.3940 \| 1.1807 \| \| No log \| 6.5854 \| 270 \| 1.5438 \| 0.2611 \| 1.5438 \| 1.2425 \| \| No log \| 6.6341 \| 272 \| 1.6434 \| 0.2110 \| 1.6434 \| 1.2819 \| \| No log \| 6.6829 \| 274 \| 1.6874 \| 0.1752 \| 1.6874 \| 1.2990 \| \| No log \| 6.7317 \| 276 \| 1.5767 \| 0.2117 \| 1.5767 \| 1.2557 \| \| No log \| 6.7805 \| 278 \| 1.4146 \| 0.1486 \| 1.4146 \| 1.1893 \| \| No log \| 6.8293 \| 280 \| 1.2838 \| 0.1142 \| 1.2838 \| 1.1330 \| \| No log \| 6.8780 \| 282 \| 1.2129 \| 0.0781 \| 1.2129 \| 1.1013 \| \| No log \| 6.9268 \| 284 \| 1.2424 \| 0.1142 \| 1.2424 \| 1.1146 \| \| No log \| 6.9756 \| 286 \| 1.3719 \| 0.2424 \| 1.3719 \| 1.1713 \| \| No log \| 7.0244 \| 288 \| 1.5464 \| 0.2437 \| 1.5464 \| 1.2435 \| \| No log \| 7.0732 \| 290 \| 1.7671 \| 0.2007 \| 1.7671 \| 1.3293 \| \| No log \| 7.1220 \| 292 \| 1.8037 \| 0.1688 \| 1.8037 \| 1.3430 \| \| No log \| 7.1707 \| 294 \| 1.7056 \| 0.2058 \| 1.7056 \| 1.3060 \| \| No log \| 7.2195 \| 296 \| 1.5468 \| 0.2239 \| 1.5468 \| 1.2437 \| \| No log \| 7.2683 \| 298 \| 1.3254 \| 0.2126 \| 1.3254 \| 1.1513 \| \| No log \| 7.3171 \| 300 \| 1.2248 \| 0.1142 \| 1.2248 \| 1.1067 \| \| No log \| 7.3659 \| 302 \| 1.2181 \| 0.1024 \| 1.2181 \| 1.1037 \| \| No log \| 7.4146 \| 304 \| 1.3199 \| 0.2752 \| 1.3199 \| 1.1489 \| \| No log \| 7.4634 \| 306 \| 1.4566 \| 0.2869 \| 1.4566 \| 1.2069 \| \| No log \| 7.5122 \| 308 \| 1.5103 \| 0.2906 \| 1.5103 \| 1.2289 \| \| No log \| 7.5610 \| 310 \| 1.5202 \| 0.2733 \| 1.5202 \| 1.2330 \| \| No log \| 7.6098 \| 312 \| 1.4770 \| 0.3177 \| 1.4770 \| 1.2153 \| \| No log \| 7.6585 \| 314 \| 1.4998 \| 0.3205 \| 1.4998 \| 1.2247 \| \| No log \| 7.7073 \| 316 \| 1.5688 \| 0.2874 \| 1.5688 \| 1.2525 \| \| No log \| 7.7561 \| 318 \| 1.6757 \| 0.2681 \| 1.6757 \| 1.2945 \| \| No log \| 7.8049 \| 320 \| 1.6044 \| 0.2482 \| 1.6044 \| 1.2667 \| \| No log \| 7.8537 \| 322 \| 1.4328 \| 0.2522 \| 1.4328 \| 1.1970 \| \| No log \| 7.9024 \| 324 \| 1.2480 \| 0.1744 \| 1.2480 \| 1.1171 \| \| No log \| 7.9512 \| 326 \| 1.1255 \| 0.0401 \| 1.1255 \| 1.0609 \| \| No log \| 8.0 \| 328 \| 1.0764 \| 0.0781 \| 1.0764 \| 1.0375 \| \| No log \| 8.0488 \| 330 \| 1.1043 \| 0.1744 \| 1.1043 \| 1.0508 \| \| No log \| 8.0976 \| 332 \| 1.2622 \| 0.3018 \| 1.2622 \| 1.1235 \| \| No log \| 8.1463 \| 334 \| 1.5303 \| 0.2606 \| 1.5303 \| 1.2370 \| \| No log \| 8.1951 \| 336 \| 1.7740 \| 0.3024 \| 1.7740 \| 1.3319 \| \| No log \| 8.2439 \| 338 \| 1.8534 \| 0.2988 \| 1.8534 \| 1.3614 \| \| No log \| 8.2927 \| 340 \| 1.7466 \| 0.2967 \| 1.7466 \| 1.3216 \| \| No log \| 8.3415 \| 342 \| 1.5761 \| 0.2940 \| 1.5761 \| 1.2554 \| \| No log \| 8.3902 \| 344 \| 1.4872 \| 0.2793 \| 1.4872 \| 1.2195 \| \| No log \| 8.4390 \| 346 \| 1.4299 \| 0.2665 \| 1.4299 \| 1.1958 \| \| No log \| 8.4878 \| 348 \| 1.4240 \| 0.2665 \| 1.4240 \| 1.1933 \| \| No log \| 8.5366 \| 350 \| 1.4802 \| 0.2474 \| 1.4802 \| 1.2166 \| \| No log \| 8.5854 \| 352 \| 1.4316 \| 0.2709 \| 1.4316 \| 1.1965 \| \| No log \| 8.6341 \| 354 \| 1.3518 \| 0.2665 \| 1.3518 \| 1.1627 \| \| No log \| 8.6829 \| 356 \| 1.4224 \| 0.2568 \| 1.4224 \| 1.1926 \| \| No log \| 8.7317 \| 358 \| 1.4891 \| 0.3117 \| 1.4891 \| 1.2203 \| \| No log \| 8.7805 \| 360 \| 1.4446 \| 0.2709 \| 1.4446 \| 1.2019 \| \| No log \| 8.8293 \| 362 \| 1.3826 \| 0.2665 \| 1.3826 \| 1.1759 \| \| No log \| 8.8780 \| 364 \| 1.3559 \| 0.2665 \| 1.3559 \| 1.1644 \| \| No log \| 8.9268 \| 366 \| 1.3007 \| 0.1814 \| 1.3007 \| 1.1405 \| \| No log \| 8.9756 \| 368 \| 1.3466 \| 0.2372 \| 1.3466 \| 1.1604 \| \| No log \| 9.0244 \| 370 \| 1.4908 \| 0.2793 \| 1.4908 \| 1.2210 \| \| No log \| 9.0732 \| 372 \| 1.5425 \| 0.3052 \| 1.5425 \| 1.2420 \| \| No log \| 9.1220 \| 374 \| 1.4719 \| 0.2752 \| 1.4719 \| 1.2132 \| \| No log \| 9.1707 \| 376 \| 1.3776 \| 0.2126 \| 1.3776 \| 1.1737 \| \| No log \| 9.2195 \| 378 \| 1.3444 \| 0.1814 \| 1.3444 \| 1.1595 \| \| No log \| 9.2683 \| 380 \| 1.3261 \| 0.1814 \| 1.3261 \| 1.1516 \| \| No log \| 9.3171 \| 382 \| 1.3477 \| 0.2126 \| 1.3477 \| 1.1609 \| \| No log \| 9.3659 \| 384 \| 1.3544 \| 0.2126 \| 1.3544 \| 1.1638 \| \| No log \| 9.4146 \| 386 \| 1.3364 \| 0.2126 \| 1.3364 \| 1.1560 \| \| No log \| 9.4634 \| 388 \| 1.3280 \| 0.2126 \| 1.3280 \| 1.1524 \| \| No log \| 9.5122 \| 390 \| 1.3023 \| 0.2424 \| 1.3023 \| 1.1412 \| \| No log \| 9.5610 \| 392 \| 1.3131 \| 0.2944 \| 1.3131 \| 1.1459 \| \| No log \| 9.6098 \| 394 \| 1.3688 \| 0.2982 \| 1.3688 \| 1.1700 \| \| No log \| 9.6585 \| 396 \| 1.4065 \| 0.3018 \| 1.4065 \| 1.1860 \| \| No log \| 9.7073 \| 398 \| 1.4039 \| 0.3018 \| 1.4039 \| 1.1849 \| \| No log \| 9.7561 \| 400 \| 1.3519 \| 0.2982 \| 1.3519 \| 1.1627 \| \| No log \| 9.8049 \| 402 \| 1.3142 \| 0.2665 \| 1.3142 \| 1.1464 \| \| No log \| 9.8537 \| 404 \| 1.3158 \| 0.2665 \| 1.3158 \| 1.1471 \| \| No log \| 9.9024 \| 406 \| 1.3620 \| 0.2665 \| 1.3620 \| 1.1671 \| \| No log \| 9.9512 \| 408 \| 1.4741 \| 0.2832 \| 1.4741 \| 1.2141 \| \| No log \| 10.0 \| 410 \| 1.4998 \| 0.2832 \| 1.4998 \| 1.2247 \| \| No log \| 10.0488 \| 412 \| 1.5107 \| 0.2832 \| 1.5107 \| 1.2291 \| \| No log \| 10.0976 \| 414 \| 1.5279 \| 0.2832 \| 1.5279 \| 1.2361 \| \| No log \| 10.1463 \| 416 \| 1.5454 \| 0.2793 \| 1.5454 \| 1.2431 \| \| No log \| 10.1951 \| 418 \| 1.5071 \| 0.2793 \| 1.5071 \| 1.2276 \| \| No log \| 10.2439 \| 420 \| 1.4512 \| 0.2982 \| 1.4512 \| 1.2047 \| \| No log \| 10.2927 \| 422 \| 1.3641 \| 0.2982 \| 1.3641 \| 1.1680 \| \| No log \| 10.3415 \| 424 \| 1.2933 \| 0.2709 \| 1.2933 \| 1.1372 \| \| No log \| 10.3902 \| 426 \| 1.2873 \| 0.2424 \| 1.2873 \| 1.1346 \| \| No log \| 10.4390 \| 428 \| 1.3117 \| 0.2424 \| 1.3117 \| 1.1453 \| \| No log \| 10.4878 \| 430 \| 1.3510 \| 0.2424 \| 1.3510 \| 1.1623 \| \| No log \| 10.5366 \| 432 \| 1.4179 \| 0.2239 \| 1.4179 \| 1.1907 \| \| No log \| 10.5854 \| 434 \| 1.4578 \| 0.2832 \| 1.4578 \| 1.2074 \| \| No log \| 10.6341 \| 436 \| 1.4536 \| 0.2568 \| 1.4536 \| 1.2057 \| \| No log \| 10.6829 \| 438 \| 1.4186 \| 0.2424 \| 1.4186 \| 1.1911 \| \| No log \| 10.7317 \| 440 \| 1.4318 \| 0.2126 \| 1.4318 \| 1.1966 \| \| No log \| 10.7805 \| 442 \| 1.4254 \| 0.2126 \| 1.4254 \| 1.1939 \| \| No log \| 10.8293 \| 444 \| 1.4542 \| 0.2126 \| 1.4542 \| 1.2059 \| \| No log \| 10.8780 \| 446 \| 1.4625 \| 0.2239 \| 1.4625 \| 1.2094 \| \| No log \| 10.9268 \| 448 \| 1.5026 \| 0.1880 \| 1.5026 \| 1.2258 \| \| No log \| 10.9756 \| 450 \| 1.4913 \| 0.1562 \| 1.4913 \| 1.2212 \| \| No log \| 11.0244 \| 452 \| 1.5060 \| 0.1024 \| 1.5060 \| 1.2272 \| \| No log \| 11.0732 \| 454 \| 1.5108 \| 0.1744 \| 1.5108 \| 1.2291 \| \| No log \| 11.1220 \| 456 \| 1.5158 \| 0.2004 \| 1.5158 \| 1.2312 \| \| No log \| 11.1707 \| 458 \| 1.5287 \| 0.2522 \| 1.5287 \| 1.2364 \| \| No log \| 11.2195 \| 460 \| 1.4799 \| 0.2832 \| 1.4799 \| 1.2165 \| \| No log \| 11.2683 \| 462 \| 1.3876 \| 0.2982 \| 1.3876 \| 1.1780 \| \| No log \| 11.3171 \| 464 \| 1.2905 \| 0.3072 \| 1.2905 \| 1.1360 \| \| No log \| 11.3659 \| 466 \| 1.2601 \| 0.3072 \| 1.2601 \| 1.1226 \| \| No log \| 11.4146 \| 468 \| 1.3082 \| 0.2709 \| 1.3082 \| 1.1438 \| \| No log \| 11.4634 \| 470 \| 1.3823 \| 0.2982 \| 1.3823 \| 1.1757 \| \| No log \| 11.5122 \| 472 \| 1.4097 \| 0.2832 \| 1.4097 \| 1.1873 \| \| No log \| 11.5610 \| 474 \| 1.3280 \| 0.2982 \| 1.3280 \| 1.1524 \| \| No log \| 11.6098 \| 476 \| 1.2140 \| 0.3460 \| 1.2140 \| 1.1018 \| \| No log \| 11.6585 \| 478 \| 1.1924 \| 0.2730 \| 1.1924 \| 1.0920 \| \| No log \| 11.7073 \| 480 \| 1.2291 \| 0.2837 \| 1.2291 \| 1.1086 \| \| No log \| 11.7561 \| 482 \| 1.3147 \| 0.2869 \| 1.3147 \| 1.1466 \| \| No log \| 11.8049 \| 484 \| 1.3998 \| 0.2694 \| 1.3998 \| 1.1831 \| \| No log \| 11.8537 \| 486 \| 1.4945 \| 0.3429 \| 1.4945 \| 1.2225 \| \| No log \| 11.9024 \| 488 \| 1.4541 \| 0.3429 \| 1.4540 \| 1.2058 \| \| No log \| 11.9512 \| 490 \| 1.3182 \| 0.3086 \| 1.3182 \| 1.1481 \| \| No log \| 12.0 \| 492 \| 1.1752 \| 0.2877 \| 1.1752 \| 1.0841 \| \| No log \| 12.0488 \| 494 \| 1.1071 \| 0.3355 \| 1.1071 \| 1.0522 \| \| No log \| 12.0976 \| 496 \| 1.1069 \| 0.3231 \| 1.1069 \| 1.0521 \| \| No log \| 12.1463 \| 498 \| 1.1869 \| 0.3052 \| 1.1869 \| 1.0895 \| \| 0.3045 \| 12.1951 \| 500 \| 1.3081 \| 0.3329 \| 1.3081 \| 1.1437 \| \| 0.3045 \| 12.2439 \| 502 \| 1.4234 \| 0.3429 \| 1.4234 \| 1.1931 \| \| 0.3045 \| 12.2927 \| 504 \| 1.4174 \| 0.3329 \| 1.4174 \| 1.1906 \| \| 0.3045 \| 12.3415 \| 506 \| 1.3559 \| 0.3086 \| 1.3559 \| 1.1644 \| \| 0.3045 \| 12.3902 \| 508 \| 1.2393 \| 0.2126 \| 1.2393 \| 1.1132 \| \| 0.3045 \| 12.4390 \| 510 \| 1.2199 \| 0.1814 \| 1.2199 \| 1.1045 \| \| 0.3045 \| 12.4878 \| 512 \| 1.2639 \| 0.1814 \| 1.2639 \| 1.1242 \| ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1
VERSIL91/13889a03-e443-4bcb-a2ab-46cfe5ea650a	VERSIL91	2025-01-20T23:08:56Z	6	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2-1.5B-Instruct", "base_model:adapter:unsloth/Qwen2-1.5B-Instruct", "license:apache-2.0", "region:us" ]	null	2025-01-20T23:08:51Z	--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2-1.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml accelerate_config: dynamo_backend: inductor mixed_precision: bf16 num_machines: 1 num_processes: auto use_cpu: false adapter: lora base_model: unsloth/Qwen2-1.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - dd06633aceb12410_train_data.json ds_type: json format: custom path: /workspace/input_data/dd06633aceb12410_train_data.json type: field_instruction: tests field_output: prompt format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 16 gradient_checkpointing: true group_by_length: false hub_model_id: null hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lora_target_modules: - q_proj - v_proj lr_scheduler: cosine max_memory: 0: 70GiB max_steps: 50 micro_batch_size: 2 mlflow_experiment_name: /tmp/dd06633aceb12410_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true quantization_config: llm_int8_enable_fp32_cpu_offload: true load_in_8bit: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer torch_compile: true train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 370ef635-02c6-4a8f-be9e-f46f2205d9d9 This model is a fine-tuned version of [unsloth/Qwen2-1.5B-Instruct](https://huggingface.co/unsloth/Qwen2-1.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 16 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 2 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.7619 \| 1 \| nan \| \| 0.0 \| 1.5238 \| 2 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
John6666/pancake-mix-illustrious-sdxl	John6666	2025-01-20T23:08:37Z	67	0	diffusers	[ "diffusers", "safetensors", "text-to-image", "stable-diffusion", "stable-diffusion-xl", "anime", "girls", "characters", "illustrious", "en", "base_model:OnomaAIResearch/Illustrious-xl-early-release-v0", "base_model:finetune:OnomaAIResearch/Illustrious-xl-early-release-v0", "license:other", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionXLPipeline", "region:us" ]	text-to-image	2025-01-20T23:02:05Z	--- license: other license_name: faipl-1.0-sd license_link: https://freedevproject.org/faipl-1.0-sd/ language: - en library_name: diffusers pipeline_tag: text-to-image tags: - text-to-image - stable-diffusion - stable-diffusion-xl - anime - girls - characters - illustrious base_model: OnomaAIResearch/Illustrious-xl-early-release-v0 --- Original model is [here](https://civitai.com/models/896658?modelVersionId=1308210). This model created by [Sukizou](https://civitai.com/user/Sukizou).
andrewmalk/kitsman	andrewmalk	2025-01-20T23:08:11Z	65	0	diffusers	[ "diffusers", "flux", "lora", "replicate", "text-to-image", "en", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "license:other", "region:us" ]	text-to-image	2025-01-17T05:27:16Z	--- license: other license_name: flux-1-dev-non-commercial-license license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md language: - en tags: - flux - diffusers - lora - replicate base_model: "black-forest-labs/FLUX.1-dev" pipeline_tag: text-to-image # widget: # - text: >- # prompt # output: # url: https://... instance_prompt: okitsman --- # Kitsman <Gallery /> Trained on Replicate using: https://replicate.com/ostris/flux-dev-lora-trainer/train ## Trigger words You should use `okitsman` to trigger the image generation. ## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers) ```py from diffusers import AutoPipelineForText2Image import torch pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda') pipeline.load_lora_weights('andrewmalk/kitsman', weight_name='lora.safetensors') image = pipeline('your prompt').images[0] ``` For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)
lesso02/1a5098f7-cebd-4eeb-a0f1-0753bde57ace	lesso02	2025-01-20T23:07:05Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/CodeLlama-13b-hf-flash", "base_model:adapter:NousResearch/CodeLlama-13b-hf-flash", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:52:47Z	--- library_name: peft base_model: NousResearch/CodeLlama-13b-hf-flash tags: - axolotl - generated_from_trainer model-index: - name: 1a5098f7-cebd-4eeb-a0f1-0753bde57ace results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/CodeLlama-13b-hf-flash bf16: true chat_template: llama3 datasets: - data_files: - 9ff4e3b24bf3b2a4_train_data.json ds_type: json format: custom path: /workspace/input_data/9ff4e3b24bf3b2a4_train_data.json type: field_input: sentence1 field_instruction: phrase1 field_output: sentence2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso02/1a5098f7-cebd-4eeb-a0f1-0753bde57ace hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/9ff4e3b24bf3b2a4_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 1a5098f7-cebd-4eeb-a0f1-0753bde57ace This model is a fine-tuned version of [NousResearch/CodeLlama-13b-hf-flash](https://huggingface.co/NousResearch/CodeLlama-13b-hf-flash) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.4527 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 16.5546 \| 0.0011 \| 1 \| 4.0087 \| \| 14.6587 \| 0.0057 \| 5 \| 3.9323 \| \| 11.4912 \| 0.0114 \| 10 \| 2.9548 \| \| 11.1484 \| 0.0171 \| 15 \| 2.5602 \| \| 9.782 \| 0.0229 \| 20 \| 2.4786 \| \| 8.6338 \| 0.0286 \| 25 \| 2.4527 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
mrHunghddddd/71740281-efde-40d2-bc98-1a144c2a49c5	mrHunghddddd	2025-01-20T23:06:46Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:Vikhrmodels/Vikhr-7B-instruct_0.4", "base_model:adapter:Vikhrmodels/Vikhr-7B-instruct_0.4", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:22:29Z	--- library_name: peft base_model: Vikhrmodels/Vikhr-7B-instruct_0.4 tags: - axolotl - generated_from_trainer model-index: - name: 71740281-efde-40d2-bc98-1a144c2a49c5 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Vikhrmodels/Vikhr-7B-instruct_0.4 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - cae8a8291b672052_train_data.json ds_type: json format: custom path: /workspace/input_data/cae8a8291b672052_train_data.json type: field_input: poem_meter field_instruction: poem_title field_output: poem_verses format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: mrHunghddddd/71740281-efde-40d2-bc98-1a144c2a49c5 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/cae8a8291b672052_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: d2d62167-223e-438b-b1e7-02a477624b1a wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: d2d62167-223e-438b-b1e7-02a477624b1a warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 71740281-efde-40d2-bc98-1a144c2a49c5 This model is a fine-tuned version of [Vikhrmodels/Vikhr-7B-instruct_0.4](https://huggingface.co/Vikhrmodels/Vikhr-7B-instruct_0.4) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.7143 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.7516 \| 0.1941 \| 200 \| 1.7143 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
MayBashendy/ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k14_task5_organization	MayBashendy	2025-01-20T23:05:46Z	7	0	transformers	[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-classification	2025-01-20T15:09:27Z	--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k14_task5_organization results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k14_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.6862 - Qwk: 0.5472 - Mse: 0.6862 - Rmse: 0.8284 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Qwk \| Mse \| Rmse \| \|:-------------:\|:-------:\|:----:\|:---------------:\|:-------:\|:------:\|:------:\| \| No log \| 0.0435 \| 2 \| 3.8865 \| -0.0294 \| 3.8865 \| 1.9714 \| \| No log \| 0.0870 \| 4 \| 2.0120 \| 0.0142 \| 2.0120 \| 1.4184 \| \| No log \| 0.1304 \| 6 \| 1.7353 \| -0.0458 \| 1.7353 \| 1.3173 \| \| No log \| 0.1739 \| 8 \| 1.2631 \| 0.0380 \| 1.2631 \| 1.1239 \| \| No log \| 0.2174 \| 10 \| 1.1664 \| -0.0032 \| 1.1664 \| 1.0800 \| \| No log \| 0.2609 \| 12 \| 1.1818 \| 0.0380 \| 1.1818 \| 1.0871 \| \| No log \| 0.3043 \| 14 \| 1.2235 \| 0.0380 \| 1.2235 \| 1.1061 \| \| No log \| 0.3478 \| 16 \| 1.2637 \| 0.0380 \| 1.2637 \| 1.1242 \| \| No log \| 0.3913 \| 18 \| 1.2607 \| 0.0760 \| 1.2607 \| 1.1228 \| \| No log \| 0.4348 \| 20 \| 1.2646 \| 0.1142 \| 1.2646 \| 1.1246 \| \| No log \| 0.4783 \| 22 \| 1.1968 \| 0.1910 \| 1.1968 \| 1.0940 \| \| No log \| 0.5217 \| 24 \| 1.1106 \| 0.1805 \| 1.1106 \| 1.0539 \| \| No log \| 0.5652 \| 26 \| 1.1018 \| 0.1493 \| 1.1018 \| 1.0497 \| \| No log \| 0.6087 \| 28 \| 1.1995 \| 0.0999 \| 1.1995 \| 1.0952 \| \| No log \| 0.6522 \| 30 \| 1.1395 \| 0.1832 \| 1.1395 \| 1.0675 \| \| No log \| 0.6957 \| 32 \| 1.1605 \| 0.0436 \| 1.1605 \| 1.0772 \| \| No log \| 0.7391 \| 34 \| 1.2550 \| 0.0883 \| 1.2550 \| 1.1203 \| \| No log \| 0.7826 \| 36 \| 1.3800 \| 0.0639 \| 1.3800 \| 1.1747 \| \| No log \| 0.8261 \| 38 \| 1.2540 \| 0.0998 \| 1.2540 \| 1.1198 \| \| No log \| 0.8696 \| 40 \| 1.1479 \| 0.1389 \| 1.1479 \| 1.0714 \| \| No log \| 0.9130 \| 42 \| 1.2217 \| 0.1028 \| 1.2217 \| 1.1053 \| \| No log \| 0.9565 \| 44 \| 1.5665 \| 0.0389 \| 1.5665 \| 1.2516 \| \| No log \| 1.0 \| 46 \| 1.6655 \| 0.0516 \| 1.6655 \| 1.2905 \| \| No log \| 1.0435 \| 48 \| 1.4103 \| 0.0598 \| 1.4103 \| 1.1876 \| \| No log \| 1.0870 \| 50 \| 1.0909 \| 0.1821 \| 1.0909 \| 1.0444 \| \| No log \| 1.1304 \| 52 \| 0.9701 \| 0.3117 \| 0.9701 \| 0.9849 \| \| No log \| 1.1739 \| 54 \| 0.9537 \| 0.3414 \| 0.9537 \| 0.9766 \| \| No log \| 1.2174 \| 56 \| 0.9618 \| 0.3562 \| 0.9618 \| 0.9807 \| \| No log \| 1.2609 \| 58 \| 0.9734 \| 0.3557 \| 0.9734 \| 0.9866 \| \| No log \| 1.3043 \| 60 \| 0.9510 \| 0.3562 \| 0.9510 \| 0.9752 \| \| No log \| 1.3478 \| 62 \| 0.9381 \| 0.3733 \| 0.9381 \| 0.9686 \| \| No log \| 1.3913 \| 64 \| 0.9339 \| 0.3014 \| 0.9339 \| 0.9664 \| \| No log \| 1.4348 \| 66 \| 0.9494 \| 0.2935 \| 0.9494 \| 0.9743 \| \| No log \| 1.4783 \| 68 \| 0.9506 \| 0.2391 \| 0.9506 \| 0.9750 \| \| No log \| 1.5217 \| 70 \| 0.9589 \| 0.2114 \| 0.9589 \| 0.9792 \| \| No log \| 1.5652 \| 72 \| 0.9002 \| 0.3014 \| 0.9002 \| 0.9488 \| \| No log \| 1.6087 \| 74 \| 0.8838 \| 0.3414 \| 0.8838 \| 0.9401 \| \| No log \| 1.6522 \| 76 \| 0.8590 \| 0.3817 \| 0.8590 \| 0.9268 \| \| No log \| 1.6957 \| 78 \| 0.8336 \| 0.3519 \| 0.8336 \| 0.9130 \| \| No log \| 1.7391 \| 80 \| 0.8431 \| 0.4 \| 0.8431 \| 0.9182 \| \| No log \| 1.7826 \| 82 \| 0.8430 \| 0.3981 \| 0.8430 \| 0.9181 \| \| No log \| 1.8261 \| 84 \| 0.8405 \| 0.3537 \| 0.8405 \| 0.9168 \| \| No log \| 1.8696 \| 86 \| 0.8561 \| 0.5155 \| 0.8561 \| 0.9252 \| \| No log \| 1.9130 \| 88 \| 0.8821 \| 0.5174 \| 0.8821 \| 0.9392 \| \| No log \| 1.9565 \| 90 \| 0.8903 \| 0.4459 \| 0.8903 \| 0.9435 \| \| No log \| 2.0 \| 92 \| 1.0073 \| 0.3972 \| 1.0073 \| 1.0036 \| \| No log \| 2.0435 \| 94 \| 1.0262 \| 0.3734 \| 1.0262 \| 1.0130 \| \| No log \| 2.0870 \| 96 \| 0.8925 \| 0.5107 \| 0.8925 \| 0.9447 \| \| No log \| 2.1304 \| 98 \| 0.8793 \| 0.4979 \| 0.8793 \| 0.9377 \| \| No log \| 2.1739 \| 100 \| 1.0192 \| 0.4416 \| 1.0192 \| 1.0096 \| \| No log \| 2.2174 \| 102 \| 0.9693 \| 0.4318 \| 0.9693 \| 0.9845 \| \| No log \| 2.2609 \| 104 \| 0.8455 \| 0.5363 \| 0.8455 \| 0.9195 \| \| No log \| 2.3043 \| 106 \| 0.8263 \| 0.5450 \| 0.8263 \| 0.9090 \| \| No log \| 2.3478 \| 108 \| 0.7896 \| 0.4813 \| 0.7896 \| 0.8886 \| \| No log \| 2.3913 \| 110 \| 0.7701 \| 0.4733 \| 0.7701 \| 0.8776 \| \| No log \| 2.4348 \| 112 \| 0.7638 \| 0.4984 \| 0.7638 \| 0.8740 \| \| No log \| 2.4783 \| 114 \| 0.7713 \| 0.5128 \| 0.7713 \| 0.8782 \| \| No log \| 2.5217 \| 116 \| 0.7180 \| 0.5635 \| 0.7180 \| 0.8474 \| \| No log \| 2.5652 \| 118 \| 0.7116 \| 0.4787 \| 0.7116 \| 0.8436 \| \| No log \| 2.6087 \| 120 \| 0.7141 \| 0.5044 \| 0.7141 \| 0.8450 \| \| No log \| 2.6522 \| 122 \| 0.7626 \| 0.6386 \| 0.7626 \| 0.8733 \| \| No log \| 2.6957 \| 124 \| 0.7743 \| 0.6435 \| 0.7743 \| 0.8800 \| \| No log \| 2.7391 \| 126 \| 0.7699 \| 0.5796 \| 0.7699 \| 0.8774 \| \| No log \| 2.7826 \| 128 \| 0.8402 \| 0.4283 \| 0.8402 \| 0.9166 \| \| No log \| 2.8261 \| 130 \| 1.0326 \| 0.3363 \| 1.0326 \| 1.0161 \| \| No log \| 2.8696 \| 132 \| 1.0287 \| 0.3761 \| 1.0287 \| 1.0143 \| \| No log \| 2.9130 \| 134 \| 0.8748 \| 0.4470 \| 0.8748 \| 0.9353 \| \| No log \| 2.9565 \| 136 \| 0.7884 \| 0.5009 \| 0.7884 \| 0.8879 \| \| No log \| 3.0 \| 138 \| 0.7676 \| 0.5570 \| 0.7676 \| 0.8761 \| \| No log \| 3.0435 \| 140 \| 0.7662 \| 0.5797 \| 0.7662 \| 0.8753 \| \| No log \| 3.0870 \| 142 \| 0.7939 \| 0.5103 \| 0.7939 \| 0.8910 \| \| No log \| 3.1304 \| 144 \| 0.8795 \| 0.4455 \| 0.8795 \| 0.9378 \| \| No log \| 3.1739 \| 146 \| 0.8895 \| 0.4935 \| 0.8895 \| 0.9431 \| \| No log \| 3.2174 \| 148 \| 0.8184 \| 0.5528 \| 0.8184 \| 0.9046 \| \| No log \| 3.2609 \| 150 \| 0.8132 \| 0.4819 \| 0.8132 \| 0.9018 \| \| No log \| 3.3043 \| 152 \| 0.8196 \| 0.4615 \| 0.8196 \| 0.9053 \| \| No log \| 3.3478 \| 154 \| 0.8178 \| 0.4419 \| 0.8178 \| 0.9043 \| \| No log \| 3.3913 \| 156 \| 0.8523 \| 0.4570 \| 0.8523 \| 0.9232 \| \| No log \| 3.4348 \| 158 \| 0.8042 \| 0.4158 \| 0.8042 \| 0.8968 \| \| No log \| 3.4783 \| 160 \| 0.8430 \| 0.3160 \| 0.8430 \| 0.9182 \| \| No log \| 3.5217 \| 162 \| 0.9556 \| 0.3401 \| 0.9556 \| 0.9776 \| \| No log \| 3.5652 \| 164 \| 0.8477 \| 0.3160 \| 0.8477 \| 0.9207 \| \| No log \| 3.6087 \| 166 \| 0.7804 \| 0.5345 \| 0.7804 \| 0.8834 \| \| No log \| 3.6522 \| 168 \| 0.8543 \| 0.4806 \| 0.8543 \| 0.9243 \| \| No log \| 3.6957 \| 170 \| 0.9146 \| 0.2865 \| 0.9146 \| 0.9563 \| \| No log \| 3.7391 \| 172 \| 0.9504 \| 0.2291 \| 0.9504 \| 0.9749 \| \| No log \| 3.7826 \| 174 \| 0.9597 \| 0.2591 \| 0.9597 \| 0.9796 \| \| No log \| 3.8261 \| 176 \| 0.9069 \| 0.3445 \| 0.9069 \| 0.9523 \| \| No log \| 3.8696 \| 178 \| 0.9335 \| 0.3811 \| 0.9335 \| 0.9662 \| \| No log \| 3.9130 \| 180 \| 0.8494 \| 0.3804 \| 0.8494 \| 0.9216 \| \| No log \| 3.9565 \| 182 \| 0.8333 \| 0.4576 \| 0.8333 \| 0.9129 \| \| No log \| 4.0 \| 184 \| 0.8048 \| 0.4730 \| 0.8048 \| 0.8971 \| \| No log \| 4.0435 \| 186 \| 0.7876 \| 0.4461 \| 0.7876 \| 0.8875 \| \| No log \| 4.0870 \| 188 \| 0.8073 \| 0.3719 \| 0.8073 \| 0.8985 \| \| No log \| 4.1304 \| 190 \| 0.7829 \| 0.4128 \| 0.7829 \| 0.8848 \| \| No log \| 4.1739 \| 192 \| 0.7779 \| 0.3996 \| 0.7779 \| 0.8820 \| \| No log \| 4.2174 \| 194 \| 0.7958 \| 0.3184 \| 0.7958 \| 0.8921 \| \| No log \| 4.2609 \| 196 \| 0.7889 \| 0.3676 \| 0.7889 \| 0.8882 \| \| No log \| 4.3043 \| 198 \| 0.7770 \| 0.5163 \| 0.7770 \| 0.8815 \| \| No log \| 4.3478 \| 200 \| 0.7898 \| 0.5766 \| 0.7898 \| 0.8887 \| \| No log \| 4.3913 \| 202 \| 0.7020 \| 0.5874 \| 0.7020 \| 0.8378 \| \| No log \| 4.4348 \| 204 \| 0.7554 \| 0.5618 \| 0.7554 \| 0.8692 \| \| No log \| 4.4783 \| 206 \| 0.7159 \| 0.5379 \| 0.7159 \| 0.8461 \| \| No log \| 4.5217 \| 208 \| 0.7211 \| 0.5330 \| 0.7211 \| 0.8492 \| \| No log \| 4.5652 \| 210 \| 0.8239 \| 0.4902 \| 0.8239 \| 0.9077 \| \| No log \| 4.6087 \| 212 \| 0.7487 \| 0.5498 \| 0.7487 \| 0.8653 \| \| No log \| 4.6522 \| 214 \| 0.7070 \| 0.4898 \| 0.7070 \| 0.8409 \| \| No log \| 4.6957 \| 216 \| 0.7224 \| 0.5213 \| 0.7224 \| 0.8500 \| \| No log \| 4.7391 \| 218 \| 0.8452 \| 0.5414 \| 0.8452 \| 0.9194 \| \| No log \| 4.7826 \| 220 \| 0.9413 \| 0.4574 \| 0.9413 \| 0.9702 \| \| No log \| 4.8261 \| 222 \| 0.8087 \| 0.5231 \| 0.8087 \| 0.8993 \| \| No log \| 4.8696 \| 224 \| 0.7436 \| 0.6007 \| 0.7436 \| 0.8623 \| \| No log \| 4.9130 \| 226 \| 0.7462 \| 0.5902 \| 0.7462 \| 0.8638 \| \| No log \| 4.9565 \| 228 \| 0.7683 \| 0.5763 \| 0.7683 \| 0.8765 \| \| No log \| 5.0 \| 230 \| 0.8623 \| 0.5020 \| 0.8623 \| 0.9286 \| \| No log \| 5.0435 \| 232 \| 0.8875 \| 0.4681 \| 0.8875 \| 0.9421 \| \| No log \| 5.0870 \| 234 \| 0.8004 \| 0.5234 \| 0.8004 \| 0.8947 \| \| No log \| 5.1304 \| 236 \| 0.7856 \| 0.4691 \| 0.7856 \| 0.8864 \| \| No log \| 5.1739 \| 238 \| 0.7886 \| 0.4918 \| 0.7886 \| 0.8880 \| \| No log \| 5.2174 \| 240 \| 0.7940 \| 0.5117 \| 0.7940 \| 0.8911 \| \| No log \| 5.2609 \| 242 \| 0.8967 \| 0.4560 \| 0.8967 \| 0.9469 \| \| No log \| 5.3043 \| 244 \| 0.9099 \| 0.4987 \| 0.9099 \| 0.9539 \| \| No log \| 5.3478 \| 246 \| 0.7790 \| 0.4410 \| 0.7790 \| 0.8826 \| \| No log \| 5.3913 \| 248 \| 0.7415 \| 0.5248 \| 0.7415 \| 0.8611 \| \| No log \| 5.4348 \| 250 \| 0.7559 \| 0.4565 \| 0.7559 \| 0.8694 \| \| No log \| 5.4783 \| 252 \| 0.7639 \| 0.4261 \| 0.7639 \| 0.8740 \| \| No log \| 5.5217 \| 254 \| 0.7792 \| 0.4494 \| 0.7792 \| 0.8827 \| \| No log \| 5.5652 \| 256 \| 0.7704 \| 0.4251 \| 0.7704 \| 0.8777 \| \| No log \| 5.6087 \| 258 \| 0.7568 \| 0.4269 \| 0.7568 \| 0.8700 \| \| No log \| 5.6522 \| 260 \| 0.7523 \| 0.4313 \| 0.7523 \| 0.8674 \| \| No log \| 5.6957 \| 262 \| 0.7222 \| 0.4787 \| 0.7222 \| 0.8498 \| \| No log \| 5.7391 \| 264 \| 0.7149 \| 0.5463 \| 0.7149 \| 0.8455 \| \| No log \| 5.7826 \| 266 \| 0.7193 \| 0.6160 \| 0.7193 \| 0.8481 \| \| No log \| 5.8261 \| 268 \| 0.7691 \| 0.5439 \| 0.7691 \| 0.8770 \| \| No log \| 5.8696 \| 270 \| 0.7113 \| 0.6617 \| 0.7113 \| 0.8434 \| \| No log \| 5.9130 \| 272 \| 0.6821 \| 0.6476 \| 0.6821 \| 0.8259 \| \| No log \| 5.9565 \| 274 \| 0.6982 \| 0.6528 \| 0.6982 \| 0.8356 \| \| No log \| 6.0 \| 276 \| 0.7793 \| 0.5318 \| 0.7793 \| 0.8828 \| \| No log \| 6.0435 \| 278 \| 0.7484 \| 0.5439 \| 0.7484 \| 0.8651 \| \| No log \| 6.0870 \| 280 \| 0.7456 \| 0.5470 \| 0.7456 \| 0.8635 \| \| No log \| 6.1304 \| 282 \| 0.7213 \| 0.5740 \| 0.7213 \| 0.8493 \| \| No log \| 6.1739 \| 284 \| 0.7244 \| 0.4873 \| 0.7244 \| 0.8511 \| \| No log \| 6.2174 \| 286 \| 0.7118 \| 0.5644 \| 0.7118 \| 0.8437 \| \| No log \| 6.2609 \| 288 \| 0.7316 \| 0.4135 \| 0.7316 \| 0.8553 \| \| No log \| 6.3043 \| 290 \| 0.7192 \| 0.4984 \| 0.7192 \| 0.8481 \| \| No log \| 6.3478 \| 292 \| 0.6941 \| 0.5163 \| 0.6941 \| 0.8331 \| \| No log \| 6.3913 \| 294 \| 0.7154 \| 0.6128 \| 0.7154 \| 0.8458 \| \| No log \| 6.4348 \| 296 \| 0.7172 \| 0.5618 \| 0.7172 \| 0.8469 \| \| No log \| 6.4783 \| 298 \| 0.6939 \| 0.5680 \| 0.6939 \| 0.8330 \| \| No log \| 6.5217 \| 300 \| 0.6980 \| 0.5877 \| 0.6980 \| 0.8354 \| \| No log \| 6.5652 \| 302 \| 0.7417 \| 0.5052 \| 0.7417 \| 0.8612 \| \| No log \| 6.6087 \| 304 \| 0.7484 \| 0.5067 \| 0.7484 \| 0.8651 \| \| No log \| 6.6522 \| 306 \| 0.7230 \| 0.4879 \| 0.7230 \| 0.8503 \| \| No log \| 6.6957 \| 308 \| 0.7159 \| 0.5002 \| 0.7159 \| 0.8461 \| \| No log \| 6.7391 \| 310 \| 0.7043 \| 0.5357 \| 0.7043 \| 0.8392 \| \| No log \| 6.7826 \| 312 \| 0.6960 \| 0.5060 \| 0.6960 \| 0.8343 \| \| No log \| 6.8261 \| 314 \| 0.6992 \| 0.5066 \| 0.6992 \| 0.8362 \| \| No log \| 6.8696 \| 316 \| 0.6768 \| 0.5809 \| 0.6768 \| 0.8227 \| \| No log \| 6.9130 \| 318 \| 0.6808 \| 0.5329 \| 0.6808 \| 0.8251 \| \| No log \| 6.9565 \| 320 \| 0.6908 \| 0.5671 \| 0.6908 \| 0.8312 \| \| No log \| 7.0 \| 322 \| 0.7242 \| 0.5263 \| 0.7242 \| 0.8510 \| \| No log \| 7.0435 \| 324 \| 0.7106 \| 0.5459 \| 0.7106 \| 0.8430 \| \| No log \| 7.0870 \| 326 \| 0.8062 \| 0.4708 \| 0.8062 \| 0.8979 \| \| No log \| 7.1304 \| 328 \| 0.8057 \| 0.4708 \| 0.8057 \| 0.8976 \| \| No log \| 7.1739 \| 330 \| 0.7553 \| 0.5046 \| 0.7553 \| 0.8691 \| \| No log \| 7.2174 \| 332 \| 0.7268 \| 0.5094 \| 0.7268 \| 0.8525 \| \| No log \| 7.2609 \| 334 \| 0.7145 \| 0.5582 \| 0.7145 \| 0.8453 \| \| No log \| 7.3043 \| 336 \| 0.7106 \| 0.5475 \| 0.7106 \| 0.8430 \| \| No log \| 7.3478 \| 338 \| 0.7219 \| 0.5446 \| 0.7219 \| 0.8497 \| \| No log \| 7.3913 \| 340 \| 0.7267 \| 0.5129 \| 0.7267 \| 0.8525 \| \| No log \| 7.4348 \| 342 \| 0.7339 \| 0.4461 \| 0.7339 \| 0.8567 \| \| No log \| 7.4783 \| 344 \| 0.7684 \| 0.4641 \| 0.7684 \| 0.8766 \| \| No log \| 7.5217 \| 346 \| 0.7471 \| 0.4641 \| 0.7471 \| 0.8644 \| \| No log \| 7.5652 \| 348 \| 0.7191 \| 0.5352 \| 0.7191 \| 0.8480 \| \| No log \| 7.6087 \| 350 \| 0.7407 \| 0.5953 \| 0.7407 \| 0.8606 \| \| No log \| 7.6522 \| 352 \| 0.7185 \| 0.5683 \| 0.7185 \| 0.8477 \| \| No log \| 7.6957 \| 354 \| 0.7478 \| 0.5476 \| 0.7478 \| 0.8647 \| \| No log \| 7.7391 \| 356 \| 0.7586 \| 0.5197 \| 0.7586 \| 0.8710 \| \| No log \| 7.7826 \| 358 \| 0.7467 \| 0.4996 \| 0.7467 \| 0.8641 \| \| No log \| 7.8261 \| 360 \| 0.8009 \| 0.4388 \| 0.8009 \| 0.8949 \| \| No log \| 7.8696 \| 362 \| 0.8052 \| 0.3269 \| 0.8052 \| 0.8973 \| \| No log \| 7.9130 \| 364 \| 0.7747 \| 0.4279 \| 0.7747 \| 0.8802 \| \| No log \| 7.9565 \| 366 \| 0.7503 \| 0.4660 \| 0.7503 \| 0.8662 \| \| No log \| 8.0 \| 368 \| 0.7182 \| 0.4918 \| 0.7182 \| 0.8475 \| \| No log \| 8.0435 \| 370 \| 0.6882 \| 0.5120 \| 0.6882 \| 0.8296 \| \| No log \| 8.0870 \| 372 \| 0.6722 \| 0.6187 \| 0.6722 \| 0.8199 \| \| No log \| 8.1304 \| 374 \| 0.7175 \| 0.6081 \| 0.7175 \| 0.8470 \| \| No log \| 8.1739 \| 376 \| 0.7358 \| 0.6071 \| 0.7358 \| 0.8578 \| \| No log \| 8.2174 \| 378 \| 0.7982 \| 0.5398 \| 0.7982 \| 0.8934 \| \| No log \| 8.2609 \| 380 \| 0.7700 \| 0.6218 \| 0.7700 \| 0.8775 \| \| No log \| 8.3043 \| 382 \| 0.6887 \| 0.5463 \| 0.6887 \| 0.8299 \| \| No log \| 8.3478 \| 384 \| 0.8329 \| 0.4508 \| 0.8329 \| 0.9127 \| \| No log \| 8.3913 \| 386 \| 0.9849 \| 0.5184 \| 0.9849 \| 0.9924 \| \| No log \| 8.4348 \| 388 \| 0.9149 \| 0.4854 \| 0.9149 \| 0.9565 \| \| No log \| 8.4783 \| 390 \| 0.7384 \| 0.5012 \| 0.7384 \| 0.8593 \| \| No log \| 8.5217 \| 392 \| 0.7104 \| 0.5473 \| 0.7104 \| 0.8428 \| \| No log \| 8.5652 \| 394 \| 0.8558 \| 0.4894 \| 0.8558 \| 0.9251 \| \| No log \| 8.6087 \| 396 \| 0.8252 \| 0.4894 \| 0.8252 \| 0.9084 \| \| No log \| 8.6522 \| 398 \| 0.7126 \| 0.5186 \| 0.7126 \| 0.8441 \| \| No log \| 8.6957 \| 400 \| 0.6932 \| 0.5432 \| 0.6932 \| 0.8326 \| \| No log \| 8.7391 \| 402 \| 0.7075 \| 0.4968 \| 0.7075 \| 0.8412 \| \| No log \| 8.7826 \| 404 \| 0.7033 \| 0.5432 \| 0.7033 \| 0.8387 \| \| No log \| 8.8261 \| 406 \| 0.7082 \| 0.5536 \| 0.7082 \| 0.8415 \| \| No log \| 8.8696 \| 408 \| 0.7080 \| 0.4760 \| 0.7080 \| 0.8414 \| \| No log \| 8.9130 \| 410 \| 0.6967 \| 0.4760 \| 0.6967 \| 0.8347 \| \| No log \| 8.9565 \| 412 \| 0.6681 \| 0.5822 \| 0.6681 \| 0.8174 \| \| No log \| 9.0 \| 414 \| 0.6506 \| 0.6046 \| 0.6506 \| 0.8066 \| \| No log \| 9.0435 \| 416 \| 0.6532 \| 0.6219 \| 0.6532 \| 0.8082 \| \| No log \| 9.0870 \| 418 \| 0.6514 \| 0.6219 \| 0.6514 \| 0.8071 \| \| No log \| 9.1304 \| 420 \| 0.6505 \| 0.5644 \| 0.6505 \| 0.8066 \| \| No log \| 9.1739 \| 422 \| 0.6544 \| 0.5886 \| 0.6544 \| 0.8090 \| \| No log \| 9.2174 \| 424 \| 0.6797 \| 0.5597 \| 0.6797 \| 0.8245 \| \| No log \| 9.2609 \| 426 \| 0.7490 \| 0.5137 \| 0.7490 \| 0.8655 \| \| No log \| 9.3043 \| 428 \| 0.7794 \| 0.4898 \| 0.7794 \| 0.8828 \| \| No log \| 9.3478 \| 430 \| 0.7201 \| 0.5400 \| 0.7201 \| 0.8486 \| \| No log \| 9.3913 \| 432 \| 0.6806 \| 0.5432 \| 0.6806 \| 0.8250 \| \| No log \| 9.4348 \| 434 \| 0.6772 \| 0.5432 \| 0.6772 \| 0.8229 \| \| No log \| 9.4783 \| 436 \| 0.6891 \| 0.5089 \| 0.6891 \| 0.8301 \| \| No log \| 9.5217 \| 438 \| 0.7677 \| 0.5428 \| 0.7677 \| 0.8762 \| \| No log \| 9.5652 \| 440 \| 0.7609 \| 0.5451 \| 0.7609 \| 0.8723 \| \| No log \| 9.6087 \| 442 \| 0.6901 \| 0.5570 \| 0.6901 \| 0.8307 \| \| No log \| 9.6522 \| 444 \| 0.6739 \| 0.5188 \| 0.6739 \| 0.8209 \| \| No log \| 9.6957 \| 446 \| 0.6775 \| 0.5074 \| 0.6775 \| 0.8231 \| \| No log \| 9.7391 \| 448 \| 0.6750 \| 0.5516 \| 0.6750 \| 0.8216 \| \| No log \| 9.7826 \| 450 \| 0.7319 \| 0.5279 \| 0.7319 \| 0.8555 \| \| No log \| 9.8261 \| 452 \| 0.7593 \| 0.5331 \| 0.7593 \| 0.8714 \| \| No log \| 9.8696 \| 454 \| 0.6942 \| 0.5098 \| 0.6942 \| 0.8332 \| \| No log \| 9.9130 \| 456 \| 0.6676 \| 0.5771 \| 0.6676 \| 0.8171 \| \| No log \| 9.9565 \| 458 \| 0.6685 \| 0.5783 \| 0.6685 \| 0.8176 \| \| No log \| 10.0 \| 460 \| 0.6652 \| 0.6307 \| 0.6652 \| 0.8156 \| \| No log \| 10.0435 \| 462 \| 0.7046 \| 0.5395 \| 0.7046 \| 0.8394 \| \| No log \| 10.0870 \| 464 \| 0.7331 \| 0.5470 \| 0.7331 \| 0.8562 \| \| No log \| 10.1304 \| 466 \| 0.7124 \| 0.4494 \| 0.7124 \| 0.8440 \| \| No log \| 10.1739 \| 468 \| 0.7124 \| 0.4893 \| 0.7124 \| 0.8441 \| \| No log \| 10.2174 \| 470 \| 0.7132 \| 0.4923 \| 0.7132 \| 0.8445 \| \| No log \| 10.2609 \| 472 \| 0.7016 \| 0.5415 \| 0.7016 \| 0.8376 \| \| No log \| 10.3043 \| 474 \| 0.6915 \| 0.5554 \| 0.6915 \| 0.8316 \| \| No log \| 10.3478 \| 476 \| 0.6783 \| 0.6177 \| 0.6783 \| 0.8236 \| \| No log \| 10.3913 \| 478 \| 0.7239 \| 0.5862 \| 0.7239 \| 0.8508 \| \| No log \| 10.4348 \| 480 \| 0.7312 \| 0.5958 \| 0.7312 \| 0.8551 \| \| No log \| 10.4783 \| 482 \| 0.6749 \| 0.6198 \| 0.6749 \| 0.8215 \| \| No log \| 10.5217 \| 484 \| 0.6417 \| 0.6335 \| 0.6417 \| 0.8011 \| \| No log \| 10.5652 \| 486 \| 0.6504 \| 0.5084 \| 0.6504 \| 0.8065 \| \| No log \| 10.6087 \| 488 \| 0.6547 \| 0.4968 \| 0.6547 \| 0.8091 \| \| No log \| 10.6522 \| 490 \| 0.6450 \| 0.5529 \| 0.6450 \| 0.8031 \| \| No log \| 10.6957 \| 492 \| 0.6673 \| 0.6073 \| 0.6673 \| 0.8169 \| \| No log \| 10.7391 \| 494 \| 0.6924 \| 0.5973 \| 0.6924 \| 0.8321 \| \| No log \| 10.7826 \| 496 \| 0.6905 \| 0.6109 \| 0.6905 \| 0.8310 \| \| No log \| 10.8261 \| 498 \| 0.6965 \| 0.5833 \| 0.6965 \| 0.8346 \| \| 0.2902 \| 10.8696 \| 500 \| 0.7077 \| 0.6209 \| 0.7077 \| 0.8412 \| \| 0.2902 \| 10.9130 \| 502 \| 0.6775 \| 0.6147 \| 0.6775 \| 0.8231 \| \| 0.2902 \| 10.9565 \| 504 \| 0.6453 \| 0.6500 \| 0.6453 \| 0.8033 \| \| 0.2902 \| 11.0 \| 506 \| 0.6393 \| 0.6175 \| 0.6393 \| 0.7996 \| \| 0.2902 \| 11.0435 \| 508 \| 0.6427 \| 0.6057 \| 0.6427 \| 0.8017 \| \| 0.2902 \| 11.0870 \| 510 \| 0.6740 \| 0.5932 \| 0.6740 \| 0.8209 \| \| 0.2902 \| 11.1304 \| 512 \| 0.7058 \| 0.5654 \| 0.7058 \| 0.8401 \| \| 0.2902 \| 11.1739 \| 514 \| 0.6862 \| 0.5472 \| 0.6862 \| 0.8284 \| ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1
mrhunghd/3ec0da79-0eab-4cb1-a6c3-1edba33530b8	mrhunghd	2025-01-20T23:02:42Z	6	0	peft	[ "peft", "safetensors", "opt", "axolotl", "generated_from_trainer", "base_model:facebook/opt-125m", "base_model:adapter:facebook/opt-125m", "license:other", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:58:34Z	--- library_name: peft license: other base_model: facebook/opt-125m tags: - axolotl - generated_from_trainer model-index: - name: 3ec0da79-0eab-4cb1-a6c3-1edba33530b8 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: facebook/opt-125m bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 1b096fb12091d0a7_train_data.json ds_type: json format: custom path: /workspace/input_data/1b096fb12091d0a7_train_data.json type: field_instruction: problem field_output: solution format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: mrhunghd/3ec0da79-0eab-4cb1-a6c3-1edba33530b8 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/1b096fb12091d0a7_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 2594ef14-2fe6-455c-8347-c1d0fb26863f wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 2594ef14-2fe6-455c-8347-c1d0fb26863f warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 3ec0da79-0eab-4cb1-a6c3-1edba33530b8 This model is a fine-tuned version of [facebook/opt-125m](https://huggingface.co/facebook/opt-125m) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.1761 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 9.3655 \| 0.2159 \| 200 \| 2.1761 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
nblinh63/f1b226a4-cdd1-4308-ac2d-1fa4ebe01843	nblinh63	2025-01-20T23:02:37Z	6	0	peft	[ "peft", "safetensors", "opt", "axolotl", "generated_from_trainer", "base_model:facebook/opt-125m", "base_model:adapter:facebook/opt-125m", "license:other", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:58:31Z	--- library_name: peft license: other base_model: facebook/opt-125m tags: - axolotl - generated_from_trainer model-index: - name: f1b226a4-cdd1-4308-ac2d-1fa4ebe01843 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: facebook/opt-125m bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 1b096fb12091d0a7_train_data.json ds_type: json format: custom path: /workspace/input_data/1b096fb12091d0a7_train_data.json type: field_instruction: problem field_output: solution format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nblinh63/f1b226a4-cdd1-4308-ac2d-1fa4ebe01843 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/1b096fb12091d0a7_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 2594ef14-2fe6-455c-8347-c1d0fb26863f wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 2594ef14-2fe6-455c-8347-c1d0fb26863f warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # f1b226a4-cdd1-4308-ac2d-1fa4ebe01843 This model is a fine-tuned version of [facebook/opt-125m](https://huggingface.co/facebook/opt-125m) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.1806 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 9.3858 \| 0.2159 \| 200 \| 2.1806 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
dzanbek/f58f03f7-8fc0-41ef-9d07-5565794dc71c	dzanbek	2025-01-20T23:02:31Z	9	0	peft	[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2-7B-Instruct", "base_model:adapter:unsloth/Qwen2-7B-Instruct", "license:apache-2.0", "region:us" ]	null	2025-01-20T22:22:46Z	--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2-7B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: f58f03f7-8fc0-41ef-9d07-5565794dc71c results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2-7B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 6e60a538f672529c_train_data.json ds_type: json format: custom path: /workspace/input_data/6e60a538f672529c_train_data.json type: field_input: communityName field_instruction: label field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: dzanbek/f58f03f7-8fc0-41ef-9d07-5565794dc71c hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 78GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/6e60a538f672529c_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 3eb92360-4e77-4cc9-9ffa-0e03d7ea7423 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 3eb92360-4e77-4cc9-9ffa-0e03d7ea7423 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # f58f03f7-8fc0-41ef-9d07-5565794dc71c This model is a fine-tuned version of [unsloth/Qwen2-7B-Instruct](https://huggingface.co/unsloth/Qwen2-7B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0001 \| 1 \| nan \| \| 0.0 \| 0.0004 \| 5 \| nan \| \| 0.0 \| 0.0008 \| 10 \| nan \| \| 0.0 \| 0.0012 \| 15 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
mrHungddddh/0ff89366-d444-4953-ad52-fc6a503d92cb	mrHungddddh	2025-01-20T23:01:12Z	6	0	peft	[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/Yarn-Mistral-7b-64k", "base_model:adapter:NousResearch/Yarn-Mistral-7b-64k", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T21:58:35Z	--- library_name: peft license: apache-2.0 base_model: NousResearch/Yarn-Mistral-7b-64k tags: - axolotl - generated_from_trainer model-index: - name: 0ff89366-d444-4953-ad52-fc6a503d92cb results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/Yarn-Mistral-7b-64k bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 5f5f73d0b6f6fe1d_train_data.json ds_type: json format: custom path: /workspace/input_data/5f5f73d0b6f6fe1d_train_data.json type: field_input: messages field_instruction: system field_output: reference format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: mrHungddddh/0ff89366-d444-4953-ad52-fc6a503d92cb hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/5f5f73d0b6f6fe1d_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 6a440f8b-4ef2-40c5-aaae-529ca715837e wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 6a440f8b-4ef2-40c5-aaae-529ca715837e warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 0ff89366-d444-4953-ad52-fc6a503d92cb This model is a fine-tuned version of [NousResearch/Yarn-Mistral-7b-64k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-64k) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.4651 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 5.1264 \| 0.0263 \| 200 \| 1.4651 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
VERSIL91/b9a1e459-4362-4834-8b97-f53708a182cd	VERSIL91	2025-01-20T23:00:32Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:huggyllama/llama-7b", "base_model:adapter:huggyllama/llama-7b", "license:other", "region:us" ]	null	2025-01-20T22:53:20Z	--- library_name: peft license: other base_model: huggyllama/llama-7b tags: - axolotl - generated_from_trainer model-index: - name: b9a1e459-4362-4834-8b97-f53708a182cd results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml accelerate_config: dynamo_backend: inductor mixed_precision: bf16 num_machines: 1 num_processes: auto use_cpu: false adapter: lora base_model: huggyllama/llama-7b bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 804c65db1d2351c5_train_data.json ds_type: json format: custom path: /workspace/input_data/804c65db1d2351c5_train_data.json type: field_instruction: sentence1 field_output: sentence2 format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 16 gradient_checkpointing: true group_by_length: false hub_model_id: VERSIL91/b9a1e459-4362-4834-8b97-f53708a182cd hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lora_target_modules: - q_proj - v_proj lr_scheduler: cosine max_memory: 0: 70GiB max_steps: 50 micro_batch_size: 2 mlflow_experiment_name: /tmp/804c65db1d2351c5_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true quantization_config: llm_int8_enable_fp32_cpu_offload: true load_in_8bit: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer torch_compile: true train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: b9a1e459-4362-4834-8b97-f53708a182cd wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: b9a1e459-4362-4834-8b97-f53708a182cd warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # b9a1e459-4362-4834-8b97-f53708a182cd This model is a fine-tuned version of [huggyllama/llama-7b](https://huggingface.co/huggyllama/llama-7b) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 16 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 50 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.0 \| 0.0010 \| 1 \| nan \| \| 0.0 \| 0.0126 \| 13 \| nan \| \| 0.0 \| 0.0252 \| 26 \| nan \| \| 0.0 \| 0.0377 \| 39 \| nan \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
mradermacher/FastApply-32B-Instruct-GGUF	mradermacher	2025-01-20T23:00:06Z	302	0	transformers	[ "transformers", "gguf", "text-generation-inference", "unsloth", "qwen2", "trl", "en", "base_model:tabnine/FastApply-32B-Instruct", "base_model:quantized:tabnine/FastApply-32B-Instruct", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]	null	2025-01-20T20:47:56Z	--- base_model: tabnine/FastApply-32B-Instruct language: - en library_name: transformers license: apache-2.0 quantized_by: mradermacher tags: - text-generation-inference - transformers - unsloth - qwen2 - trl --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/tabnine/FastApply-32B-Instruct <!-- provided-files --> weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q2_K.gguf) \| Q2_K \| 12.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q3_K_S.gguf) \| Q3_K_S \| 14.5 \| \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q3_K_M.gguf) \| Q3_K_M \| 16.0 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q3_K_L.gguf) \| Q3_K_L \| 17.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.IQ4_XS.gguf) \| IQ4_XS \| 18.0 \| \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q4_K_S.gguf) \| Q4_K_S \| 18.9 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q4_K_M.gguf) \| Q4_K_M \| 20.0 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q5_K_S.gguf) \| Q5_K_S \| 22.7 \| \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q5_K_M.gguf) \| Q5_K_M \| 23.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q6_K.gguf) \| Q6_K \| 27.0 \| very good quality \| \| [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q8_0.gguf) \| Q8_0 \| 34.9 \| fast, best quality \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
nhunglaaaaaaa/31738ab5-8e3e-4c3d-a223-8d795b64609e	nhunglaaaaaaa	2025-01-20T22:58:35Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:heegyu/WizardVicuna-open-llama-3b-v2", "base_model:adapter:heegyu/WizardVicuna-open-llama-3b-v2", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]	null	2025-01-20T22:05:54Z	--- library_name: peft license: apache-2.0 base_model: heegyu/WizardVicuna-open-llama-3b-v2 tags: - axolotl - generated_from_trainer model-index: - name: 31738ab5-8e3e-4c3d-a223-8d795b64609e results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: heegyu/WizardVicuna-open-llama-3b-v2 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 212d5e0168a48c19_train_data.json ds_type: json format: custom path: /workspace/input_data/212d5e0168a48c19_train_data.json type: field_instruction: context_en field_output: question_en format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nhunglaaaaaaa/31738ab5-8e3e-4c3d-a223-8d795b64609e hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/212d5e0168a48c19_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 1e1fd096-ba6e-478c-9e4b-c08c22fc3c74 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 1e1fd096-ba6e-478c-9e4b-c08c22fc3c74 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 31738ab5-8e3e-4c3d-a223-8d795b64609e This model is a fine-tuned version of [heegyu/WizardVicuna-open-llama-3b-v2](https://huggingface.co/heegyu/WizardVicuna-open-llama-3b-v2) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.8120 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.0691 \| 0.0026 \| 200 \| 0.8120 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
dimasik1987/0b3322d1-4a39-4579-a195-f5358131b723	dimasik1987	2025-01-20T22:58:05Z	6	0	peft	[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/Yarn-Mistral-7b-64k", "base_model:adapter:NousResearch/Yarn-Mistral-7b-64k", "license:apache-2.0", "region:us" ]	null	2025-01-20T21:58:35Z	--- library_name: peft license: apache-2.0 base_model: NousResearch/Yarn-Mistral-7b-64k tags: - axolotl - generated_from_trainer model-index: - name: 0b3322d1-4a39-4579-a195-f5358131b723 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/Yarn-Mistral-7b-64k bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 5f5f73d0b6f6fe1d_train_data.json ds_type: json format: custom path: /workspace/input_data/5f5f73d0b6f6fe1d_train_data.json type: field_input: messages field_instruction: system field_output: reference format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: dimasik1987/0b3322d1-4a39-4579-a195-f5358131b723 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/5f5f73d0b6f6fe1d_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 6a440f8b-4ef2-40c5-aaae-529ca715837e wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 6a440f8b-4ef2-40c5-aaae-529ca715837e warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # 0b3322d1-4a39-4579-a195-f5358131b723 This model is a fine-tuned version of [NousResearch/Yarn-Mistral-7b-64k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-64k) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.8223 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| No log \| 0.0003 \| 1 \| 1.2169 \| \| 3.8291 \| 0.0013 \| 5 \| 1.0084 \| \| 3.1617 \| 0.0026 \| 10 \| 0.8943 \| \| 3.1678 \| 0.0039 \| 15 \| 0.8525 \| \| 3.2285 \| 0.0053 \| 20 \| 0.8367 \| \| 3.3153 \| 0.0066 \| 25 \| 0.8247 \| \| 3.5174 \| 0.0079 \| 30 \| 0.8223 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
lilmeaty/xfsfsfsf-4bit	lilmeaty	2025-01-20T22:57:54Z	20	0	transformers	[ "transformers", "pytorch", "safetensors", "llama", "text-generation", "conversational", "license:apache-2.0", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-01-20T07:50:13Z	--- license: apache-2.0 library_name: transformers ---
prxy5607/feb3717a-696e-4e00-8c82-54646ee762f9	prxy5607	2025-01-20T22:56:12Z	6	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:deepseek-ai/deepseek-coder-6.7b-instruct", "base_model:adapter:deepseek-ai/deepseek-coder-6.7b-instruct", "license:other", "region:us" ]	null	2025-01-20T21:36:32Z	--- library_name: peft license: other base_model: deepseek-ai/deepseek-coder-6.7b-instruct tags: - axolotl - generated_from_trainer model-index: - name: feb3717a-696e-4e00-8c82-54646ee762f9 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: deepseek-ai/deepseek-coder-6.7b-instruct bf16: true chat_template: llama3 data_processes: 16 dataset_prepared_path: null datasets: - data_files: - aaf4cd02348b6ba9_train_data.json ds_type: json format: custom path: /workspace/input_data/aaf4cd02348b6ba9_train_data.json type: field_instruction: code field_output: docstring format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto do_eval: true early_stopping_patience: 5 eval_batch_size: 4 eval_max_new_tokens: 128 eval_steps: 50 eval_table_size: null evals_per_epoch: null flash_attention: true fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: prxy5607/feb3717a-696e-4e00-8c82-54646ee762f9 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 128 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 64 lora_target_linear: true lr_scheduler: cosine max_grad_norm: 1.0 max_memory: 0: 75GB max_steps: 200 micro_batch_size: 8 mlflow_experiment_name: /tmp/aaf4cd02348b6ba9_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 50 saves_per_epoch: null sequence_len: 1024 strict: false tf32: true tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 6b80479d-4e03-4f5b-b68e-f811da024a88 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 6b80479d-4e03-4f5b-b68e-f811da024a88 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # feb3717a-696e-4e00-8c82-54646ee762f9 This model is a fine-tuned version of [deepseek-ai/deepseek-coder-6.7b-instruct](https://huggingface.co/deepseek-ai/deepseek-coder-6.7b-instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2623 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 8 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 2.6902 \| 0.0002 \| 1 \| 3.2009 \| \| 1.0728 \| 0.0092 \| 50 \| 1.3635 \| \| 1.2867 \| 0.0185 \| 100 \| 1.2835 \| \| 1.3301 \| 0.0277 \| 150 \| 1.2678 \| \| 1.3324 \| 0.0369 \| 200 \| 1.2623 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

nhunglaaaaaaa/fccd41c3-68c9-4af6-9694-bcf592003ec3

nhunglaaaaaaa

2025-01-20T23:58:40Z

8

0

peft

[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "base_model:NousResearch/Hermes-2-Pro-Mistral-7B", "base_model:adapter:NousResearch/Hermes-2-Pro-Mistral-7B", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:37:37Z

--- library_name: peft license: apache-2.0 base_model: NousResearch/Hermes-2-Pro-Mistral-7B tags: - axolotl - generated_from_trainer model-index: - name: fccd41c3-68c9-4af6-9694-bcf592003ec3 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/Hermes-2-Pro-Mistral-7B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 670127402f937a76_train_data.json ds_type: json format: custom path: /workspace/input_data/670127402f937a76_train_data.json type: field_input: content field_instruction: aspect field_output: sentiment format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nhunglaaaaaaa/fccd41c3-68c9-4af6-9694-bcf592003ec3 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/670127402f937a76_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: e13dcc00-4d7d-439f-bbb5-ed9061820333 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: e13dcc00-4d7d-439f-bbb5-ed9061820333 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # fccd41c3-68c9-4af6-9694-bcf592003ec3 This model is a fine-tuned version of [NousResearch/Hermes-2-Pro-Mistral-7B](https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2271 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.9569 | 0.6070 | 200 | 0.2271 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

ClarenceDan/eb986324-3dd5-4e2b-868e-a31fb3ad18d2

ClarenceDan

2025-01-20T23:57:26Z

8

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "region:us" ]

null

2025-01-20T23:41:06Z

--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: eb986324-3dd5-4e2b-868e-a31fb3ad18d2 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 854bca96bed40197_train_data.json ds_type: json format: custom path: /workspace/input_data/854bca96bed40197_train_data.json type: field_input: state_before field_instruction: tactic field_output: state_after format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: ClarenceDan/eb986324-3dd5-4e2b-868e-a31fb3ad18d2 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/854bca96bed40197_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: cff9d1c5-a847-4707-b347-d0451baf6b24 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: cff9d1c5-a847-4707-b347-d0451baf6b24 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # eb986324-3dd5-4e2b-868e-a31fb3ad18d2 This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.1671 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 2.1411 | 0.0000 | 1 | 2.8415 | | 0.8263 | 0.0001 | 3 | 2.8217 | | 0.3436 | 0.0002 | 6 | 2.6084 | | 0.4392 | 0.0003 | 9 | 2.1671 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

lesso11/1db4e142-5823-4549-b311-c5325f577241

lesso11

2025-01-20T23:55:46Z

8

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:48:03Z

--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: 1db4e142-5823-4549-b311-c5325f577241 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: true chat_template: llama3 datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso11/1db4e142-5823-4549-b311-c5325f577241 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 1db4e142-5823-4549-b311-c5325f577241 This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.0004 | 1 | nan | | 0.0 | 0.0021 | 5 | nan | | 0.0 | 0.0042 | 10 | nan | | 0.0 | 0.0063 | 15 | nan | | 0.0 | 0.0085 | 20 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

joboffer/c80d23f2-e3e7-4dab-a9fa-432251a8e9d7

joboffer

2025-01-20T23:54:50Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "region:us" ]

null

2025-01-20T23:48:23Z

--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: c80d23f2-e3e7-4dab-a9fa-432251a8e9d7 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: joboffer/c80d23f2-e3e7-4dab-a9fa-432251a8e9d7 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 80GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 2048 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # c80d23f2-e3e7-4dab-a9fa-432251a8e9d7 This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0004 | 1 | nan | | 0.0 | 0.0021 | 5 | nan | | 0.0 | 0.0042 | 10 | nan | | 0.0 | 0.0063 | 15 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

nadejdatarabukina/c739481e-a786-4a3a-b2b4-3cd78c486811

nadejdatarabukina

2025-01-20T23:54:08Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "region:us" ]

null

2025-01-20T23:48:13Z

--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: c739481e-a786-4a3a-b2b4-3cd78c486811 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: null eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: nadejdatarabukina/c739481e-a786-4a3a-b2b4-3cd78c486811 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 75GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 10 weight_decay: 0.01 xformers_attention: true ``` </details><br> # c739481e-a786-4a3a-b2b4-3cd78c486811 This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0004 | 1 | nan | | 0.0 | 0.0021 | 5 | nan | | 0.0 | 0.0042 | 10 | nan | | 0.0 | 0.0063 | 15 | nan | | 0.0 | 0.0085 | 20 | nan | | 0.0 | 0.0106 | 25 | nan | | 0.0 | 0.0127 | 30 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

LHRuig/aireduxreal

LHRuig

2025-01-20T23:53:34Z

22

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:53:25Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # aireduxreal <Gallery /> ## Model description aireduxreal lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/aireduxreal/tree/main) them in the Files & versions tab.

lesso09/f177f676-835b-4da0-b0b0-ede96e7966ba

lesso09

2025-01-20T23:52:14Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:princeton-nlp/Sheared-LLaMA-1.3B", "base_model:adapter:princeton-nlp/Sheared-LLaMA-1.3B", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:50:43Z

--- library_name: peft license: apache-2.0 base_model: princeton-nlp/Sheared-LLaMA-1.3B tags: - axolotl - generated_from_trainer model-index: - name: f177f676-835b-4da0-b0b0-ede96e7966ba results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: princeton-nlp/Sheared-LLaMA-1.3B bf16: true chat_template: llama3 datasets: - data_files: - 641edd60ca44ac19_train_data.json ds_type: json format: custom path: /workspace/input_data/641edd60ca44ac19_train_data.json type: field_input: tokens field_instruction: sentence field_output: corrupted format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso09/f177f676-835b-4da0-b0b0-ede96e7966ba hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/641edd60ca44ac19_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ab7df99e-418b-42ee-9175-b39bc5d0f0ce wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ab7df99e-418b-42ee-9175-b39bc5d0f0ce warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # f177f676-835b-4da0-b0b0-ede96e7966ba This model is a fine-tuned version of [princeton-nlp/Sheared-LLaMA-1.3B](https://huggingface.co/princeton-nlp/Sheared-LLaMA-1.3B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.0026 | 1 | nan | | 0.0 | 0.0130 | 5 | nan | | 0.0 | 0.0260 | 10 | nan | | 0.0 | 0.0390 | 15 | nan | | 0.0 | 0.0520 | 20 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

MayBashendy/ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k18_task5_organization

MayBashendy

2025-01-20T23:52:05Z

7

0

transformers

[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]

text-classification

2025-01-20T16:11:17Z

--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k18_task5_organization results: [] ---  # ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k18_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.0523 - Qwk: 0.4790 - Mse: 1.0523 - Rmse: 1.0258 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | Qwk | Mse | Rmse | |:-------------:|:------:|:----:|:---------------:|:-------:|:------:|:------:| | No log | 0.0345 | 2 | 3.8827 | -0.0151 | 3.8827 | 1.9705 | | No log | 0.0690 | 4 | 2.3503 | -0.0189 | 2.3503 | 1.5331 | | No log | 0.1034 | 6 | 3.3976 | -0.0343 | 3.3976 | 1.8433 | | No log | 0.1379 | 8 | 4.3908 | 0.0171 | 4.3908 | 2.0954 | | No log | 0.1724 | 10 | 2.9131 | -0.0295 | 2.9131 | 1.7068 | | No log | 0.2069 | 12 | 1.3737 | 0.0894 | 1.3737 | 1.1720 | | No log | 0.2414 | 14 | 1.0962 | 0.2416 | 1.0962 | 1.0470 | | No log | 0.2759 | 16 | 1.0484 | 0.3449 | 1.0484 | 1.0239 | | No log | 0.3103 | 18 | 1.1433 | 0.3927 | 1.1433 | 1.0692 | | No log | 0.3448 | 20 | 1.5454 | 0.1429 | 1.5454 | 1.2431 | | No log | 0.3793 | 22 | 1.9269 | 0.1296 | 1.9269 | 1.3881 | | No log | 0.4138 | 24 | 2.1738 | 0.1465 | 2.1738 | 1.4744 | | No log | 0.4483 | 26 | 2.2513 | 0.1221 | 2.2513 | 1.5004 | | No log | 0.4828 | 28 | 1.6165 | 0.2351 | 1.6165 | 1.2714 | | No log | 0.5172 | 30 | 1.1390 | 0.2560 | 1.1390 | 1.0672 | | No log | 0.5517 | 32 | 0.9792 | 0.2526 | 0.9792 | 0.9895 | | No log | 0.5862 | 34 | 1.0101 | 0.3066 | 1.0101 | 1.0050 | | No log | 0.6207 | 36 | 1.1235 | 0.4012 | 1.1235 | 1.0600 | | No log | 0.6552 | 38 | 1.1648 | 0.3542 | 1.1648 | 1.0793 | | No log | 0.6897 | 40 | 1.1022 | 0.3243 | 1.1022 | 1.0499 | | No log | 0.7241 | 42 | 1.1218 | 0.3243 | 1.1218 | 1.0591 | | No log | 0.7586 | 44 | 1.0679 | 0.2977 | 1.0679 | 1.0334 | | No log | 0.7931 | 46 | 1.0611 | 0.2567 | 1.0611 | 1.0301 | | No log | 0.8276 | 48 | 0.9578 | 0.3691 | 0.9578 | 0.9787 | | No log | 0.8621 | 50 | 0.8676 | 0.2770 | 0.8676 | 0.9315 | | No log | 0.8966 | 52 | 0.8749 | 0.2770 | 0.8749 | 0.9354 | | No log | 0.9310 | 54 | 0.8883 | 0.3498 | 0.8883 | 0.9425 | | No log | 0.9655 | 56 | 0.9563 | 0.3250 | 0.9563 | 0.9779 | | No log | 1.0 | 58 | 1.1407 | 0.3283 | 1.1407 | 1.0681 | | No log | 1.0345 | 60 | 1.0584 | 0.2711 | 1.0584 | 1.0288 | | No log | 1.0690 | 62 | 0.9629 | 0.3129 | 0.9629 | 0.9813 | | No log | 1.1034 | 64 | 1.0630 | 0.3902 | 1.0630 | 1.0310 | | No log | 1.1379 | 66 | 1.1496 | 0.2471 | 1.1496 | 1.0722 | | No log | 1.1724 | 68 | 1.1000 | 0.3024 | 1.1000 | 1.0488 | | No log | 1.2069 | 70 | 0.9757 | 0.3326 | 0.9757 | 0.9878 | | No log | 1.2414 | 72 | 0.9358 | 0.3198 | 0.9358 | 0.9674 | | No log | 1.2759 | 74 | 0.9934 | 0.1794 | 0.9934 | 0.9967 | | No log | 1.3103 | 76 | 1.0084 | 0.2188 | 1.0084 | 1.0042 | | No log | 1.3448 | 78 | 1.0067 | 0.3713 | 1.0067 | 1.0033 | | No log | 1.3793 | 80 | 0.9656 | 0.3455 | 0.9656 | 0.9827 | | No log | 1.4138 | 82 | 0.9832 | 0.3198 | 0.9832 | 0.9915 | | No log | 1.4483 | 84 | 1.0364 | 0.2795 | 1.0364 | 1.0180 | | No log | 1.4828 | 86 | 1.0609 | 0.3063 | 1.0609 | 1.0300 | | No log | 1.5172 | 88 | 1.0794 | 0.2582 | 1.0794 | 1.0389 | | No log | 1.5517 | 90 | 1.0631 | 0.3860 | 1.0631 | 1.0311 | | No log | 1.5862 | 92 | 1.0703 | 0.3915 | 1.0703 | 1.0346 | | No log | 1.6207 | 94 | 1.2684 | 0.3070 | 1.2684 | 1.1262 | | No log | 1.6552 | 96 | 1.4565 | 0.3333 | 1.4565 | 1.2068 | | No log | 1.6897 | 98 | 1.3804 | 0.3318 | 1.3804 | 1.1749 | | No log | 1.7241 | 100 | 1.1446 | 0.2837 | 1.1446 | 1.0698 | | No log | 1.7586 | 102 | 1.0397 | 0.3133 | 1.0397 | 1.0197 | | No log | 1.7931 | 104 | 1.0443 | 0.4373 | 1.0443 | 1.0219 | | No log | 1.8276 | 106 | 1.0965 | 0.4392 | 1.0965 | 1.0471 | | No log | 1.8621 | 108 | 1.0463 | 0.4206 | 1.0463 | 1.0229 | | No log | 1.8966 | 110 | 1.0460 | 0.3482 | 1.0460 | 1.0228 | | No log | 1.9310 | 112 | 1.0853 | 0.3711 | 1.0853 | 1.0418 | | No log | 1.9655 | 114 | 1.1043 | 0.3758 | 1.1043 | 1.0509 | | No log | 2.0 | 116 | 1.1332 | 0.3718 | 1.1332 | 1.0645 | | No log | 2.0345 | 118 | 1.1377 | 0.3917 | 1.1377 | 1.0666 | | No log | 2.0690 | 120 | 1.1435 | 0.3648 | 1.1435 | 1.0693 | | No log | 2.1034 | 122 | 1.1013 | 0.4232 | 1.1013 | 1.0494 | | No log | 2.1379 | 124 | 0.9759 | 0.3607 | 0.9759 | 0.9879 | | No log | 2.1724 | 126 | 0.8873 | 0.3820 | 0.8873 | 0.9420 | | No log | 2.2069 | 128 | 0.8663 | 0.3625 | 0.8663 | 0.9307 | | No log | 2.2414 | 130 | 0.8625 | 0.4089 | 0.8625 | 0.9287 | | No log | 2.2759 | 132 | 0.8833 | 0.4056 | 0.8833 | 0.9398 | | No log | 2.3103 | 134 | 0.9124 | 0.4313 | 0.9124 | 0.9552 | | No log | 2.3448 | 136 | 0.9654 | 0.3286 | 0.9654 | 0.9826 | | No log | 2.3793 | 138 | 1.1335 | 0.3687 | 1.1335 | 1.0647 | | No log | 2.4138 | 140 | 1.0551 | 0.3918 | 1.0551 | 1.0272 | | No log | 2.4483 | 142 | 1.0378 | 0.4737 | 1.0378 | 1.0187 | | No log | 2.4828 | 144 | 1.1433 | 0.4264 | 1.1433 | 1.0693 | | No log | 2.5172 | 146 | 1.0562 | 0.4503 | 1.0562 | 1.0277 | | No log | 2.5517 | 148 | 1.0218 | 0.2694 | 1.0218 | 1.0109 | | No log | 2.5862 | 150 | 0.9906 | 0.3514 | 0.9906 | 0.9953 | | No log | 2.6207 | 152 | 0.9382 | 0.3224 | 0.9382 | 0.9686 | | No log | 2.6552 | 154 | 0.9202 | 0.2941 | 0.9202 | 0.9592 | | No log | 2.6897 | 156 | 0.9203 | 0.3357 | 0.9203 | 0.9593 | | No log | 2.7241 | 158 | 0.9347 | 0.4272 | 0.9347 | 0.9668 | | No log | 2.7586 | 160 | 0.9881 | 0.4273 | 0.9881 | 0.9940 | | No log | 2.7931 | 162 | 1.0588 | 0.4309 | 1.0588 | 1.0290 | | No log | 2.8276 | 164 | 1.1410 | 0.4973 | 1.1410 | 1.0682 | | No log | 2.8621 | 166 | 1.3562 | 0.2730 | 1.3562 | 1.1646 | | No log | 2.8966 | 168 | 1.4988 | 0.2789 | 1.4988 | 1.2242 | | No log | 2.9310 | 170 | 1.3422 | 0.2230 | 1.3422 | 1.1585 | | No log | 2.9655 | 172 | 1.1816 | 0.4020 | 1.1816 | 1.0870 | | No log | 3.0 | 174 | 1.1237 | 0.3924 | 1.1237 | 1.0601 | | No log | 3.0345 | 176 | 1.1163 | 0.3843 | 1.1163 | 1.0566 | | No log | 3.0690 | 178 | 1.2763 | 0.4032 | 1.2763 | 1.1297 | | No log | 3.1034 | 180 | 1.3986 | 0.3243 | 1.3986 | 1.1826 | | No log | 3.1379 | 182 | 1.2453 | 0.3345 | 1.2453 | 1.1159 | | No log | 3.1724 | 184 | 0.9703 | 0.4161 | 0.9703 | 0.9850 | | No log | 3.2069 | 186 | 0.9294 | 0.3804 | 0.9294 | 0.9641 | | No log | 3.2414 | 188 | 0.9516 | 0.4845 | 0.9516 | 0.9755 | | No log | 3.2759 | 190 | 0.9303 | 0.4223 | 0.9303 | 0.9645 | | No log | 3.3103 | 192 | 1.0026 | 0.4196 | 1.0026 | 1.0013 | | No log | 3.3448 | 194 | 1.1127 | 0.4976 | 1.1127 | 1.0548 | | No log | 3.3793 | 196 | 1.0237 | 0.4568 | 1.0237 | 1.0118 | | No log | 3.4138 | 198 | 0.9068 | 0.3454 | 0.9068 | 0.9523 | | No log | 3.4483 | 200 | 0.8817 | 0.4118 | 0.8817 | 0.9390 | | No log | 3.4828 | 202 | 0.9082 | 0.4729 | 0.9082 | 0.9530 | | No log | 3.5172 | 204 | 0.9099 | 0.4581 | 0.9099 | 0.9539 | | No log | 3.5517 | 206 | 0.8902 | 0.4350 | 0.8902 | 0.9435 | | No log | 3.5862 | 208 | 0.9101 | 0.4215 | 0.9101 | 0.9540 | | No log | 3.6207 | 210 | 0.9443 | 0.4807 | 0.9443 | 0.9717 | | No log | 3.6552 | 212 | 0.9715 | 0.4369 | 0.9715 | 0.9856 | | No log | 3.6897 | 214 | 0.9362 | 0.3792 | 0.9362 | 0.9676 | | No log | 3.7241 | 216 | 0.9121 | 0.3742 | 0.9121 | 0.9550 | | No log | 3.7586 | 218 | 0.9766 | 0.4585 | 0.9766 | 0.9882 | | No log | 3.7931 | 220 | 1.1081 | 0.4787 | 1.1081 | 1.0527 | | No log | 3.8276 | 222 | 1.1726 | 0.4301 | 1.1726 | 1.0829 | | No log | 3.8621 | 224 | 1.0732 | 0.4585 | 1.0732 | 1.0360 | | No log | 3.8966 | 226 | 0.9207 | 0.4230 | 0.9207 | 0.9595 | | No log | 3.9310 | 228 | 0.8838 | 0.4114 | 0.8838 | 0.9401 | | No log | 3.9655 | 230 | 0.9041 | 0.4230 | 0.9041 | 0.9508 | | No log | 4.0 | 232 | 0.9400 | 0.4745 | 0.9400 | 0.9696 | | No log | 4.0345 | 234 | 0.9597 | 0.4930 | 0.9597 | 0.9796 | | No log | 4.0690 | 236 | 0.9359 | 0.4306 | 0.9359 | 0.9674 | | No log | 4.1034 | 238 | 0.8728 | 0.4069 | 0.8728 | 0.9342 | | No log | 4.1379 | 240 | 0.8607 | 0.3981 | 0.8607 | 0.9277 | | No log | 4.1724 | 242 | 0.8876 | 0.4405 | 0.8876 | 0.9421 | | No log | 4.2069 | 244 | 0.8922 | 0.4160 | 0.8922 | 0.9445 | | No log | 4.2414 | 246 | 0.9119 | 0.5073 | 0.9119 | 0.9549 | | No log | 4.2759 | 248 | 0.9036 | 0.4364 | 0.9036 | 0.9506 | | No log | 4.3103 | 250 | 0.9250 | 0.4760 | 0.9250 | 0.9618 | | No log | 4.3448 | 252 | 0.9678 | 0.4076 | 0.9678 | 0.9838 | | No log | 4.3793 | 254 | 0.9431 | 0.4063 | 0.9431 | 0.9711 | | No log | 4.4138 | 256 | 0.9360 | 0.3705 | 0.9360 | 0.9675 | | No log | 4.4483 | 258 | 0.9044 | 0.4186 | 0.9044 | 0.9510 | | No log | 4.4828 | 260 | 0.9198 | 0.3705 | 0.9198 | 0.9590 | | No log | 4.5172 | 262 | 1.0117 | 0.4058 | 1.0117 | 1.0058 | | No log | 4.5517 | 264 | 1.0365 | 0.4286 | 1.0365 | 1.0181 | | No log | 4.5862 | 266 | 0.9587 | 0.4033 | 0.9587 | 0.9791 | | No log | 4.6207 | 268 | 0.9625 | 0.3842 | 0.9625 | 0.9811 | | No log | 4.6552 | 270 | 1.0778 | 0.4405 | 1.0778 | 1.0382 | | No log | 4.6897 | 272 | 1.2736 | 0.3972 | 1.2736 | 1.1286 | | No log | 4.7241 | 274 | 1.2229 | 0.4874 | 1.2229 | 1.1059 | | No log | 4.7586 | 276 | 1.0118 | 0.4510 | 1.0118 | 1.0059 | | No log | 4.7931 | 278 | 0.8990 | 0.3640 | 0.8990 | 0.9482 | | No log | 4.8276 | 280 | 0.8903 | 0.3640 | 0.8903 | 0.9435 | | No log | 4.8621 | 282 | 0.9633 | 0.4171 | 0.9633 | 0.9815 | | No log | 4.8966 | 284 | 1.1899 | 0.4491 | 1.1899 | 1.0908 | | No log | 4.9310 | 286 | 1.1471 | 0.4487 | 1.1471 | 1.0710 | | No log | 4.9655 | 288 | 0.9659 | 0.3512 | 0.9659 | 0.9828 | | No log | 5.0 | 290 | 0.9598 | 0.3607 | 0.9598 | 0.9797 | | No log | 5.0345 | 292 | 1.0185 | 0.3629 | 1.0185 | 1.0092 | | No log | 5.0690 | 294 | 1.0927 | 0.3766 | 1.0927 | 1.0453 | | No log | 5.1034 | 296 | 1.2132 | 0.4186 | 1.2132 | 1.1015 | | No log | 5.1379 | 298 | 1.1753 | 0.3972 | 1.1753 | 1.0841 | | No log | 5.1724 | 300 | 1.0267 | 0.3436 | 1.0267 | 1.0132 | | No log | 5.2069 | 302 | 0.9602 | 0.3144 | 0.9602 | 0.9799 | | No log | 5.2414 | 304 | 0.9394 | 0.3308 | 0.9394 | 0.9692 | | No log | 5.2759 | 306 | 0.9721 | 0.3463 | 0.9721 | 0.9859 | | No log | 5.3103 | 308 | 1.0627 | 0.3972 | 1.0627 | 1.0309 | | No log | 5.3448 | 310 | 1.0346 | 0.4589 | 1.0346 | 1.0172 | | No log | 5.3793 | 312 | 0.8910 | 0.3809 | 0.8910 | 0.9439 | | No log | 5.4138 | 314 | 0.8535 | 0.3437 | 0.8535 | 0.9238 | | No log | 5.4483 | 316 | 0.8935 | 0.3447 | 0.8935 | 0.9453 | | No log | 5.4828 | 318 | 0.9043 | 0.3842 | 0.9043 | 0.9510 | | No log | 5.5172 | 320 | 1.0219 | 0.4613 | 1.0219 | 1.0109 | | No log | 5.5517 | 322 | 1.1213 | 0.4693 | 1.1213 | 1.0589 | | No log | 5.5862 | 324 | 1.0448 | 0.4615 | 1.0448 | 1.0222 | | No log | 5.6207 | 326 | 0.9554 | 0.4376 | 0.9554 | 0.9775 | | No log | 5.6552 | 328 | 0.9215 | 0.4273 | 0.9215 | 0.9600 | | No log | 5.6897 | 330 | 0.9059 | 0.3678 | 0.9059 | 0.9518 | | No log | 5.7241 | 332 | 0.9500 | 0.3728 | 0.9500 | 0.9747 | | No log | 5.7586 | 334 | 0.9382 | 0.3688 | 0.9382 | 0.9686 | | No log | 5.7931 | 336 | 0.8925 | 0.2939 | 0.8925 | 0.9447 | | No log | 5.8276 | 338 | 0.8878 | 0.4229 | 0.8878 | 0.9422 | | No log | 5.8621 | 340 | 0.8877 | 0.4013 | 0.8877 | 0.9422 | | No log | 5.8966 | 342 | 0.9140 | 0.3923 | 0.9140 | 0.9560 | | No log | 5.9310 | 344 | 0.9673 | 0.4822 | 0.9673 | 0.9835 | | No log | 5.9655 | 346 | 1.0367 | 0.4589 | 1.0367 | 1.0182 | | No log | 6.0 | 348 | 0.9746 | 0.4792 | 0.9746 | 0.9872 | | No log | 6.0345 | 350 | 0.8739 | 0.3535 | 0.8739 | 0.9348 | | No log | 6.0690 | 352 | 0.8633 | 0.3572 | 0.8633 | 0.9291 | | No log | 6.1034 | 354 | 0.8985 | 0.4677 | 0.8985 | 0.9479 | | No log | 6.1379 | 356 | 0.9457 | 0.4449 | 0.9457 | 0.9725 | | No log | 6.1724 | 358 | 1.0018 | 0.3975 | 1.0018 | 1.0009 | | No log | 6.2069 | 360 | 1.0198 | 0.3728 | 1.0198 | 1.0099 | | No log | 6.2414 | 362 | 0.9638 | 0.3624 | 0.9638 | 0.9817 | | No log | 6.2759 | 364 | 0.9300 | 0.3772 | 0.9300 | 0.9644 | | No log | 6.3103 | 366 | 0.9246 | 0.3860 | 0.9246 | 0.9616 | | No log | 6.3448 | 368 | 0.9481 | 0.3877 | 0.9481 | 0.9737 | | No log | 6.3793 | 370 | 1.0751 | 0.4497 | 1.0751 | 1.0369 | | No log | 6.4138 | 372 | 1.0104 | 0.4400 | 1.0104 | 1.0052 | | No log | 6.4483 | 374 | 0.8589 | 0.4003 | 0.8589 | 0.9267 | | No log | 6.4828 | 376 | 0.8292 | 0.5179 | 0.8292 | 0.9106 | | No log | 6.5172 | 378 | 0.8432 | 0.5071 | 0.8432 | 0.9183 | | No log | 6.5517 | 380 | 0.8499 | 0.3601 | 0.8499 | 0.9219 | | No log | 6.5862 | 382 | 1.0001 | 0.4694 | 1.0001 | 1.0001 | | No log | 6.6207 | 384 | 1.1535 | 0.4681 | 1.1535 | 1.0740 | | No log | 6.6552 | 386 | 1.1194 | 0.4592 | 1.1194 | 1.0580 | | No log | 6.6897 | 388 | 0.9916 | 0.4228 | 0.9916 | 0.9958 | | No log | 6.7241 | 390 | 0.9721 | 0.3065 | 0.9721 | 0.9859 | | No log | 6.7586 | 392 | 0.9257 | 0.3159 | 0.9257 | 0.9621 | | No log | 6.7931 | 394 | 0.8857 | 0.2416 | 0.8857 | 0.9411 | | No log | 6.8276 | 396 | 0.8897 | 0.2951 | 0.8897 | 0.9432 | | No log | 6.8621 | 398 | 0.9185 | 0.3029 | 0.9185 | 0.9584 | | No log | 6.8966 | 400 | 0.9333 | 0.3134 | 0.9333 | 0.9661 | | No log | 6.9310 | 402 | 0.9419 | 0.3207 | 0.9419 | 0.9705 | | No log | 6.9655 | 404 | 1.0028 | 0.3917 | 1.0028 | 1.0014 | | No log | 7.0 | 406 | 1.0369 | 0.3946 | 1.0369 | 1.0183 | | No log | 7.0345 | 408 | 1.0417 | 0.4116 | 1.0417 | 1.0206 | | No log | 7.0690 | 410 | 1.0446 | 0.3348 | 1.0446 | 1.0221 | | No log | 7.1034 | 412 | 1.0095 | 0.2742 | 1.0095 | 1.0047 | | No log | 7.1379 | 414 | 1.0162 | 0.3430 | 1.0162 | 1.0081 | | No log | 7.1724 | 416 | 0.9666 | 0.3404 | 0.9666 | 0.9832 | | No log | 7.2069 | 418 | 0.9279 | 0.2786 | 0.9279 | 0.9633 | | No log | 7.2414 | 420 | 0.9377 | 0.3457 | 0.9377 | 0.9683 | | No log | 7.2759 | 422 | 0.9122 | 0.3457 | 0.9122 | 0.9551 | | No log | 7.3103 | 424 | 0.8928 | 0.2899 | 0.8928 | 0.9449 | | No log | 7.3448 | 426 | 0.9117 | 0.3601 | 0.9117 | 0.9549 | | No log | 7.3793 | 428 | 0.8795 | 0.3289 | 0.8795 | 0.9378 | | No log | 7.4138 | 430 | 0.8721 | 0.3357 | 0.8721 | 0.9339 | | No log | 7.4483 | 432 | 0.8763 | 0.3224 | 0.8763 | 0.9361 | | No log | 7.4828 | 434 | 0.9084 | 0.5222 | 0.9084 | 0.9531 | | No log | 7.5172 | 436 | 0.8770 | 0.4268 | 0.8770 | 0.9365 | | No log | 7.5517 | 438 | 0.8508 | 0.3980 | 0.8508 | 0.9224 | | No log | 7.5862 | 440 | 0.9223 | 0.4205 | 0.9223 | 0.9604 | | No log | 7.6207 | 442 | 1.0067 | 0.4400 | 1.0067 | 1.0034 | | No log | 7.6552 | 444 | 0.9343 | 0.4822 | 0.9343 | 0.9666 | | No log | 7.6897 | 446 | 0.8419 | 0.4247 | 0.8419 | 0.9175 | | No log | 7.7241 | 448 | 0.8061 | 0.4463 | 0.8061 | 0.8979 | | No log | 7.7586 | 450 | 0.8104 | 0.4221 | 0.8104 | 0.9002 | | No log | 7.7931 | 452 | 0.8290 | 0.4327 | 0.8290 | 0.9105 | | No log | 7.8276 | 454 | 0.8036 | 0.4088 | 0.8036 | 0.8964 | | No log | 7.8621 | 456 | 0.7561 | 0.5431 | 0.7561 | 0.8695 | | No log | 7.8966 | 458 | 0.7412 | 0.5057 | 0.7412 | 0.8609 | | No log | 7.9310 | 460 | 0.7442 | 0.4706 | 0.7442 | 0.8627 | | No log | 7.9655 | 462 | 0.7870 | 0.4421 | 0.7870 | 0.8871 | | No log | 8.0 | 464 | 0.8435 | 0.4103 | 0.8435 | 0.9184 | | No log | 8.0345 | 466 | 0.8581 | 0.4306 | 0.8581 | 0.9263 | | No log | 8.0690 | 468 | 0.8054 | 0.4425 | 0.8054 | 0.8974 | | No log | 8.1034 | 470 | 0.7978 | 0.4444 | 0.7978 | 0.8932 | | No log | 8.1379 | 472 | 0.7923 | 0.3985 | 0.7923 | 0.8901 | | No log | 8.1724 | 474 | 0.7926 | 0.4235 | 0.7926 | 0.8903 | | No log | 8.2069 | 476 | 0.7973 | 0.4297 | 0.7973 | 0.8929 | | No log | 8.2414 | 478 | 0.7842 | 0.3967 | 0.7842 | 0.8856 | | No log | 8.2759 | 480 | 0.7854 | 0.3970 | 0.7854 | 0.8862 | | No log | 8.3103 | 482 | 0.8013 | 0.4622 | 0.8013 | 0.8952 | | No log | 8.3448 | 484 | 0.8078 | 0.4622 | 0.8078 | 0.8988 | | No log | 8.3793 | 486 | 0.7642 | 0.5328 | 0.7642 | 0.8742 | | No log | 8.4138 | 488 | 0.8224 | 0.4719 | 0.8224 | 0.9069 | | No log | 8.4483 | 490 | 1.0762 | 0.5161 | 1.0762 | 1.0374 | | No log | 8.4828 | 492 | 1.1014 | 0.5145 | 1.1014 | 1.0495 | | No log | 8.5172 | 494 | 0.9148 | 0.4810 | 0.9148 | 0.9565 | | No log | 8.5517 | 496 | 0.7722 | 0.4577 | 0.7722 | 0.8787 | | No log | 8.5862 | 498 | 0.8145 | 0.5435 | 0.8145 | 0.9025 | | 0.312 | 8.6207 | 500 | 0.8249 | 0.4903 | 0.8249 | 0.9083 | | 0.312 | 8.6552 | 502 | 0.8091 | 0.4072 | 0.8091 | 0.8995 | | 0.312 | 8.6897 | 504 | 0.8141 | 0.3455 | 0.8141 | 0.9023 | | 0.312 | 8.7241 | 506 | 0.8404 | 0.3563 | 0.8404 | 0.9167 | | 0.312 | 8.7586 | 508 | 0.9525 | 0.4156 | 0.9525 | 0.9759 | | 0.312 | 8.7931 | 510 | 1.0523 | 0.4790 | 1.0523 | 1.0258 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1

dimasik2987/c4fd30f8-9bed-4199-bfca-cc8c495b627b

dimasik2987

2025-01-20T23:52:01Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "region:us" ]

null

2025-01-20T23:48:29Z

--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: c4fd30f8-9bed-4199-bfca-cc8c495b627b results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: dimasik2987/c4fd30f8-9bed-4199-bfca-cc8c495b627b hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # c4fd30f8-9bed-4199-bfca-cc8c495b627b This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0008 | 1 | nan | | 0.0 | 0.0042 | 5 | nan | | 0.0 | 0.0085 | 10 | nan | | 0.0 | 0.0127 | 15 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

VERSIL91/8e83e5ba-b92a-4907-bf39-218caf42228a

VERSIL91

2025-01-20T23:51:49Z

8

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:HuggingFaceM4/tiny-random-LlamaForCausalLM", "base_model:adapter:HuggingFaceM4/tiny-random-LlamaForCausalLM", "region:us" ]

null

2025-01-20T23:51:46Z

--- library_name: peft base_model: HuggingFaceM4/tiny-random-LlamaForCausalLM tags: - axolotl - generated_from_trainer model-index: - name: bef97220-cdcf-4144-9f98-4582cf4a902b results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml accelerate_config: dynamo_backend: inductor mixed_precision: bf16 num_machines: 1 num_processes: auto use_cpu: false adapter: lora base_model: HuggingFaceM4/tiny-random-LlamaForCausalLM bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 7da10487b55868a6_train_data.json ds_type: json format: custom path: /workspace/input_data/7da10487b55868a6_train_data.json type: field_instruction: hyps field_output: ref format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 16 gradient_checkpointing: true group_by_length: false hub_model_id: null hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lora_target_modules: - q_proj - v_proj lr_scheduler: cosine max_memory: 0: 70GiB max_steps: 50 micro_batch_size: 2 mlflow_experiment_name: /tmp/7da10487b55868a6_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true quantization_config: llm_int8_enable_fp32_cpu_offload: true load_in_8bit: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer torch_compile: true train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: bef97220-cdcf-4144-9f98-4582cf4a902b wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: bef97220-cdcf-4144-9f98-4582cf4a902b warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # bef97220-cdcf-4144-9f98-4582cf4a902b This model is a fine-tuned version of [HuggingFaceM4/tiny-random-LlamaForCausalLM](https://huggingface.co/HuggingFaceM4/tiny-random-LlamaForCausalLM) on the None dataset. It achieves the following results on the evaluation set: - Loss: 10.3648 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 16 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 50 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 10.3772 | 0.0002 | 1 | 10.3803 | | 10.3751 | 0.0031 | 13 | 10.3758 | | 10.3722 | 0.0062 | 26 | 10.3686 | | 10.3688 | 0.0093 | 39 | 10.3648 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

kostiantynk1205/41925306-0ef0-4197-b10b-98e1dc6ca5d3

kostiantynk1205

2025-01-20T23:51:38Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2-1.5B-Instruct", "base_model:adapter:unsloth/Qwen2-1.5B-Instruct", "license:apache-2.0", "region:us" ]

null

2025-01-20T23:51:09Z

--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2-1.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 41925306-0ef0-4197-b10b-98e1dc6ca5d3 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2-1.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - dd06633aceb12410_train_data.json ds_type: json format: custom path: /workspace/input_data/dd06633aceb12410_train_data.json type: field_instruction: tests field_output: prompt format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: kostiantynk1205/41925306-0ef0-4197-b10b-98e1dc6ca5d3 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/dd06633aceb12410_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 wandb_project: Birthday-SN56-23-Gradients-On-Demand wandb_run: your_name wandb_runid: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 41925306-0ef0-4197-b10b-98e1dc6ca5d3 This model is a fine-tuned version of [unsloth/Qwen2-1.5B-Instruct](https://huggingface.co/unsloth/Qwen2-1.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 6 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.1905 | 1 | nan | | 0.0 | 0.3810 | 2 | nan | | 0.0 | 0.7619 | 4 | nan | | 0.0 | 1.1429 | 6 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

dimasik87/150164aa-8220-47b5-a42e-e3e1914c42b8

dimasik87

2025-01-20T23:50:30Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "custom_code", "base_model:NovaSearch/stella_en_1.5B_v5", "base_model:adapter:NovaSearch/stella_en_1.5B_v5", "license:mit", "region:us" ]

null

2025-01-20T23:48:18Z

--- library_name: peft license: mit base_model: dunzhang/stella_en_1.5B_v5 tags: - axolotl - generated_from_trainer model-index: - name: 150164aa-8220-47b5-a42e-e3e1914c42b8 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: dunzhang/stella_en_1.5B_v5 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 09f295eb7fd803c2_train_data.json ds_type: json format: custom path: /workspace/input_data/09f295eb7fd803c2_train_data.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: dimasik87/150164aa-8220-47b5-a42e-e3e1914c42b8 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/09f295eb7fd803c2_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: a727ab38-5312-4e68-8885-5980a4cae8a9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: a727ab38-5312-4e68-8885-5980a4cae8a9 warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # 150164aa-8220-47b5-a42e-e3e1914c42b8 This model is a fine-tuned version of [dunzhang/stella_en_1.5B_v5](https://huggingface.co/dunzhang/stella_en_1.5B_v5) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0008 | 1 | nan | | 0.0 | 0.0042 | 5 | nan | | 0.0 | 0.0085 | 10 | nan | | 0.0 | 0.0127 | 15 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

ardaspear/32cc01b9-8751-4d53-b346-be211e673aa6

ardaspear

2025-01-20T23:49:47Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:codellama/CodeLlama-7b-hf", "base_model:adapter:codellama/CodeLlama-7b-hf", "license:llama2", "region:us" ]

null

2025-01-20T22:33:09Z

--- library_name: peft license: llama2 base_model: codellama/CodeLlama-7b-hf tags: - axolotl - generated_from_trainer model-index: - name: 32cc01b9-8751-4d53-b346-be211e673aa6 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: codellama/CodeLlama-7b-hf bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 25670fa5d5514c5b_train_data.json ds_type: json format: custom path: /workspace/input_data/25670fa5d5514c5b_train_data.json type: field_input: facts field_instruction: prompt_serial field_output: hypothesis format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: ardaspear/32cc01b9-8751-4d53-b346-be211e673aa6 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 5.0e-05 load_in_4bit: false load_in_8bit: false local_rank: 0 logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 100 micro_batch_size: 8 mlflow_experiment_name: /tmp/25670fa5d5514c5b_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: techspear-hub wandb_mode: online wandb_name: ac3b7fa4-8d2e-4dc9-b139-aaeea5f132ac wandb_project: Gradients-On-Five wandb_run: your_name wandb_runid: ac3b7fa4-8d2e-4dc9-b139-aaeea5f132ac warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # 32cc01b9-8751-4d53-b346-be211e673aa6 This model is a fine-tuned version of [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.0776 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0008 | 1 | 0.7974 | | 0.7688 | 0.0076 | 9 | 0.7061 | | 0.3383 | 0.0152 | 18 | 0.3166 | | 0.174 | 0.0228 | 27 | 0.1894 | | 0.147 | 0.0305 | 36 | 0.1512 | | 0.1256 | 0.0381 | 45 | 0.1198 | | 0.1727 | 0.0457 | 54 | 0.0931 | | 0.0712 | 0.0533 | 63 | 0.0830 | | 0.0857 | 0.0609 | 72 | 0.0793 | | 0.0679 | 0.0685 | 81 | 0.0780 | | 0.039 | 0.0762 | 90 | 0.0777 | | 0.0732 | 0.0838 | 99 | 0.0776 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

tuanna08go/db6fb768-4f34-45e9-a10c-21c12445199e

tuanna08go

2025-01-20T23:49:36Z

8

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:codellama/CodeLlama-7b-hf", "base_model:adapter:codellama/CodeLlama-7b-hf", "license:llama2", "region:us" ]

null

2025-01-20T23:09:11Z

--- library_name: peft license: llama2 base_model: codellama/CodeLlama-7b-hf tags: - axolotl - generated_from_trainer model-index: - name: db6fb768-4f34-45e9-a10c-21c12445199e results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: codellama/CodeLlama-7b-hf bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 25670fa5d5514c5b_train_data.json ds_type: json format: custom path: /workspace/input_data/25670fa5d5514c5b_train_data.json type: field_input: facts field_instruction: prompt_serial field_output: hypothesis format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 5 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: tuanna08go/db6fb768-4f34-45e9-a10c-21c12445199e hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 5 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 50 micro_batch_size: 2 mlflow_experiment_name: /tmp/25670fa5d5514c5b_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ac3b7fa4-8d2e-4dc9-b139-aaeea5f132ac wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ac3b7fa4-8d2e-4dc9-b139-aaeea5f132ac warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # db6fb768-4f34-45e9-a10c-21c12445199e This model is a fine-tuned version of [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.5158 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 50 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0002 | 1 | 1.2075 | | 1.0753 | 0.0021 | 10 | 1.0732 | | 0.7292 | 0.0042 | 20 | 0.6758 | | 0.5962 | 0.0063 | 30 | 0.5538 | | 0.4682 | 0.0085 | 40 | 0.5197 | | 0.4209 | 0.0106 | 50 | 0.5158 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

charlesniswander/fruit_freshness_demo

charlesniswander

2025-01-20T23:49:26Z

28

0

null

[ "tensorboard", "safetensors", "vit", "image-classification", "pytorch", "huggingpics", "model-index", "region:us" ]

image-classification

2025-01-20T23:49:14Z

--- tags: - image-classification - pytorch - huggingpics metrics: - accuracy model-index: - name: fruit_freshness_demo results: - task: name: Image Classification type: image-classification metrics: - name: Accuracy type: accuracy value: 0.3483146131038666 --- # fruit_freshness_demo Autogenerated by HuggingPics🤗🖼️ Create your own image classifier for **anything** by running [the demo on Google Colab](https://colab.research.google.com/github/nateraw/huggingpics/blob/main/HuggingPics.ipynb). Report any issues with the demo at the [github repo](https://github.com/nateraw/huggingpics). ## Example Images #### fresh apple ![fresh apple](images/fresh_apple.jpg) #### fresh banana ![fresh banana](images/fresh_banana.jpg) #### rotten apple ![rotten apple](images/rotten_apple.jpg) #### rotten banana ![rotten banana](images/rotten_banana.jpg)

taopanda-2/a31b520a-ea45-423d-a41b-09df0cfc856c

taopanda-2

2025-01-20T23:49:24Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:06:04Z

--- license: apache-2.0 library_name: peft tags: - axolotl - generated_from_trainer base_model: Qwen/Qwen2.5-0.5B-Instruct model-index: - name: a31b520a-ea45-423d-a41b-09df0cfc856c results: [] ---  [<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto dataset_prepared_path: null datasets: - data_files: - a7412d8c8f805ddf_train_data.json ds_type: json format: custom path: a7412d8c8f805ddf_train_data.json type: field: null field_input: null field_instruction: premise field_output: hypothesis field_system: null format: null no_input_format: null system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 2 flash_attention: null fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: taopanda-2/a31b520a-ea45-423d-a41b-09df0cfc856c learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 32 lora_target_linear: true lr_scheduler: cosine micro_batch_size: 2 model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: ./outputs/lora-out/taopanda-2_ba28db81-f399-44e5-bdef-7af8dcf5a4ca pad_to_sequence_len: null resume_from_checkpoint: null sample_packing: false saves_per_epoch: 1 seed: 91813 sequence_len: 2048 special_tokens: null strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: fatcat87-taopanda wandb_log_model: null wandb_mode: online wandb_name: taopanda-2_ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_project: subnet56 wandb_runid: taopanda-2_ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_watch: null warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/fatcat87-taopanda/subnet56/runs/sk5fvbjg) # a31b520a-ea45-423d-a41b-09df0cfc856c This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.9854 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 91813 - distributed_type: multi-GPU - num_devices: 4 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - total_eval_batch_size: 8 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - num_epochs: 1 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 4.4358 | 0.0006 | 1 | 4.7686 | | 0.8153 | 0.4998 | 823 | 1.0209 | | 0.9222 | 0.9997 | 1646 | 0.9854 | ### Framework versions - PEFT 0.11.1 - Transformers 4.42.3 - Pytorch 2.3.0+cu121 - Datasets 2.19.1 - Tokenizers 0.19.1

nat-hunt/13925d41-82bd-407e-b7bd-93b7227d91e6

nat-hunt

2025-01-20T23:49:00Z

8

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:unsloth/Meta-Llama-3.1-8B", "base_model:adapter:unsloth/Meta-Llama-3.1-8B", "license:llama3.1", "region:us" ]

null

2025-01-20T23:45:49Z

--- library_name: peft license: llama3.1 base_model: unsloth/Meta-Llama-3.1-8B tags: - axolotl - generated_from_trainer model-index: - name: 13925d41-82bd-407e-b7bd-93b7227d91e6 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Meta-Llama-3.1-8B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 91e193d3dca1611f_train_data.json ds_type: json format: custom path: /workspace/input_data/91e193d3dca1611f_train_data.json type: field_input: parent_id field_instruction: role field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: nat-hunt/13925d41-82bd-407e-b7bd-93b7227d91e6 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/91e193d3dca1611f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 856a9aac-189f-40f7-b27c-c5616995b0d1 wandb_project: Birthday-SN56-25-Gradients-On-Demand wandb_run: your_name wandb_runid: 856a9aac-189f-40f7-b27c-c5616995b0d1 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 13925d41-82bd-407e-b7bd-93b7227d91e6 This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.0004 | 1 | nan | | 0.0 | 0.0011 | 3 | nan | | 0.0 | 0.0021 | 6 | nan | | 0.0 | 0.0032 | 9 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

LHRuig/doubleexpo

LHRuig

2025-01-20T23:48:34Z

9

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:48:28Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # doubleexpo <Gallery /> ## Model description doubleexpo lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/doubleexpo/tree/main) them in the Files & versions tab.

mradermacher/QwQ-56B-Ghost-i1-GGUF

mradermacher

2025-01-20T23:48:26Z

513

0

transformers

[ "transformers", "gguf", "mergekit", "merge", "uncensored", "abliterated", "chat", "en", "base_model:JackCloudman/QwQ-56B-Ghost", "base_model:quantized:JackCloudman/QwQ-56B-Ghost", "license:apache-2.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ]

null

2025-01-20T12:39:13Z

--- base_model: JackCloudman/QwQ-56B-Ghost language: - en library_name: transformers license: apache-2.0 quantized_by: mradermacher tags: - mergekit - merge - uncensored - abliterated - chat --- ## About      weighted/imatrix quants of https://huggingface.co/JackCloudman/QwQ-56B-Ghost  static quants are available at https://huggingface.co/mradermacher/QwQ-56B-Ghost-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ1_S.gguf) | i1-IQ1_S | 12.2 | for the desperate | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ1_M.gguf) | i1-IQ1_M | 13.4 | mostly desperate | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 15.3 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ2_XS.gguf) | i1-IQ2_XS | 16.9 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ2_S.gguf) | i1-IQ2_S | 17.6 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ2_M.gguf) | i1-IQ2_M | 19.2 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q2_K_S.gguf) | i1-Q2_K_S | 19.5 | very low quality | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q2_K.gguf) | i1-Q2_K | 21.0 | IQ3_XXS probably better | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 21.9 | lower quality | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ3_XS.gguf) | i1-IQ3_XS | 23.4 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q3_K_S.gguf) | i1-Q3_K_S | 24.6 | IQ3_XS probably better | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ3_S.gguf) | i1-IQ3_S | 24.6 | beats Q3_K* | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ3_M.gguf) | i1-IQ3_M | 25.3 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q3_K_M.gguf) | i1-Q3_K_M | 27.3 | IQ3_S probably better | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q3_K_L.gguf) | i1-Q3_K_L | 29.5 | IQ3_M probably better | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-IQ4_XS.gguf) | i1-IQ4_XS | 30.3 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q4_0.gguf) | i1-Q4_0 | 32.0 | fast, low quality | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q4_K_S.gguf) | i1-Q4_K_S | 32.2 | optimal size/speed/quality | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q4_K_M.gguf) | i1-Q4_K_M | 34.0 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q4_1.gguf) | i1-Q4_1 | 35.4 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q5_K_S.gguf) | i1-Q5_K_S | 38.8 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q5_K_M.gguf) | i1-Q5_K_M | 39.9 | | | [GGUF](https://huggingface.co/mradermacher/QwQ-56B-Ghost-i1-GGUF/resolve/main/QwQ-56B-Ghost.i1-Q6_K.gguf) | i1-Q6_K | 46.2 | practically like static Q6_K | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.

mradermacher/LwQ-Reasoner-10B-i1-GGUF

mradermacher

2025-01-20T23:48:26Z

580

0

transformers

[ "transformers", "gguf", "LlamaWithQuestions", "CoT", "Reasoner", "LWQ", "en", "base_model:prithivMLmods/LwQ-Reasoner-10B", "base_model:quantized:prithivMLmods/LwQ-Reasoner-10B", "license:llama3.1", "endpoints_compatible", "region:us", "imatrix" ]

null

2025-01-20T22:25:16Z

--- base_model: prithivMLmods/LwQ-Reasoner-10B language: - en library_name: transformers license: llama3.1 quantized_by: mradermacher tags: - LlamaWithQuestions - CoT - Reasoner - LWQ --- ## About      weighted/imatrix quants of https://huggingface.co/prithivMLmods/LwQ-Reasoner-10B  static quants are available at https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ1_S.gguf) | i1-IQ1_S | 2.5 | for the desperate | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ1_M.gguf) | i1-IQ1_M | 2.7 | mostly desperate | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 3.0 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 3.3 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ2_S.gguf) | i1-IQ2_S | 3.4 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ2_M.gguf) | i1-IQ2_M | 3.7 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 3.8 | very low quality | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q2_K.gguf) | i1-Q2_K | 4.0 | IQ3_XXS probably better | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 4.2 | lower quality | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 4.5 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ3_S.gguf) | i1-IQ3_S | 4.7 | beats Q3_K* | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 4.7 | IQ3_XS probably better | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ3_M.gguf) | i1-IQ3_M | 4.8 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 5.2 | IQ3_S probably better | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 5.6 | IQ3_M probably better | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 5.7 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-IQ4_NL.gguf) | i1-IQ4_NL | 6.0 | prefer IQ4_XS | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q4_0.gguf) | i1-Q4_0 | 6.0 | fast, low quality | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 6.1 | optimal size/speed/quality | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 6.4 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q4_1.gguf) | i1-Q4_1 | 6.6 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 7.2 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 7.4 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF/resolve/main/LwQ-Reasoner-10B.i1-Q6_K.gguf) | i1-Q6_K | 8.6 | practically like static Q6_K | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.

jncraton/DeepSeek-R1-Distill-Llama-8B-ct2-int8

jncraton

2025-01-20T23:47:44Z

12

0

null

[ "region:us" ]

null

2025-01-20T23:33:07Z

# DeepSeek-R1    <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- **Post-Training: Large-Scale Reinforcement Learning on the Base Model** - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- **Distillation: Smaller Models Can Be Powerful Too** - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> | **Model** | **#Total Params** | **#Activated Params** | **Context Length** | **Download** | | :------------: | :------------: | :------------: | :------------: | :------------: | | DeepSeek-R1-Zero | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) | | DeepSeek-R1 | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) | </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> | **Model** | **Base Model** | **Download** | | :------------: | :------------: | :------------: | | DeepSeek-R1-Distill-Qwen-1.5B | [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) | | DeepSeek-R1-Distill-Qwen-7B | [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) | | DeepSeek-R1-Distill-Llama-8B | [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) | | DeepSeek-R1-Distill-Qwen-14B | [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) | |DeepSeek-R1-Distill-Qwen-32B | [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) | | DeepSeek-R1-Distill-Llama-70B | [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) | </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> | Category | Benchmark (Metric) | Claude-3.5-Sonnet-1022 | GPT-4o 0513 | DeepSeek V3 | OpenAI o1-mini | OpenAI o1-1217 | DeepSeek R1 | |----------|-------------------|----------------------|------------|--------------|----------------|------------|--------------| | | Architecture | - | - | MoE | - | - | MoE | | | # Activated Params | - | - | 37B | - | - | 37B | | | # Total Params | - | - | 671B | - | - | 671B | | English | MMLU (Pass@1) | 88.3 | 87.2 | 88.5 | 85.2 | **91.8** | 90.8 | | | MMLU-Redux (EM) | 88.9 | 88.0 | 89.1 | 86.7 | - | **92.9** | | | MMLU-Pro (EM) | 78.0 | 72.6 | 75.9 | 80.3 | - | **84.0** | | | DROP (3-shot F1) | 88.3 | 83.7 | 91.6 | 83.9 | 90.2 | **92.2** | | | IF-Eval (Prompt Strict) | **86.5** | 84.3 | 86.1 | 84.8 | - | 83.3 | | | GPQA-Diamond (Pass@1) | 65.0 | 49.9 | 59.1 | 60.0 | **75.7** | 71.5 | | | SimpleQA (Correct) | 28.4 | 38.2 | 24.9 | 7.0 | **47.0** | 30.1 | | | FRAMES (Acc.) | 72.5 | 80.5 | 73.3 | 76.9 | - | **82.5** | | | AlpacaEval2.0 (LC-winrate) | 52.0 | 51.1 | 70.0 | 57.8 | - | **87.6** | | | ArenaHard (GPT-4-1106) | 85.2 | 80.4 | 85.5 | 92.0 | - | **92.3** | | Code | LiveCodeBench (Pass@1-COT) | 33.8 | 34.2 | - | 53.8 | 63.4 | **65.9** | | | Codeforces (Percentile) | 20.3 | 23.6 | 58.7 | 93.4 | **96.6** | 96.3 | | | Codeforces (Rating) | 717 | 759 | 1134 | 1820 | **2061** | 2029 | | | SWE Verified (Resolved) | **50.8** | 38.8 | 42.0 | 41.6 | 48.9 | 49.2 | | | Aider-Polyglot (Acc.) | 45.3 | 16.0 | 49.6 | 32.9 | **61.7** | 53.3 | | Math | AIME 2024 (Pass@1) | 16.0 | 9.3 | 39.2 | 63.6 | 79.2 | **79.8** | | | MATH-500 (Pass@1) | 78.3 | 74.6 | 90.2 | 90.0 | 96.4 | **97.3** | | | CNMO 2024 (Pass@1) | 13.1 | 10.8 | 43.2 | 67.6 | - | **78.8** | | Chinese | CLUEWSC (EM) | 85.4 | 87.9 | 90.9 | 89.9 | - | **92.8** | | | C-Eval (EM) | 76.7 | 76.0 | 86.5 | 68.9 | - | **91.8** | | | C-SimpleQA (Correct) | 55.4 | 58.7 | **68.0** | 40.3 | - | 63.7 | </div> ### Distilled Model Evaluation <div align="center"> | Model | AIME 2024 pass@1 | AIME 2024 cons@64 | MATH-500 pass@1 | GPQA Diamond pass@1 | LiveCodeBench pass@1 | CodeForces rating | |------------------------------------------|------------------|-------------------|-----------------|----------------------|----------------------|-------------------| | GPT-4o-0513 | 9.3 | 13.4 | 74.6 | 49.9 | 32.9 | 759 | | Claude-3.5-Sonnet-1022 | 16.0 | 26.7 | 78.3 | 65.0 | 38.9 | 717 | | o1-mini | 63.6 | 80.0 | 90.0 | 60.0 | 53.8 | **1820** | | QwQ-32B-Preview | 44.0 | 60.0 | 90.6 | 54.5 | 41.9 | 1316 | | DeepSeek-R1-Distill-Qwen-1.5B | 28.9 | 52.7 | 83.9 | 33.8 | 16.9 | 954 | | DeepSeek-R1-Distill-Qwen-7B | 55.5 | 83.3 | 92.8 | 49.1 | 37.6 | 1189 | | DeepSeek-R1-Distill-Qwen-14B | 69.7 | 80.0 | 93.9 | 59.1 | 53.1 | 1481 | | DeepSeek-R1-Distill-Qwen-32B | **72.6** | 83.3 | 94.3 | 62.1 | 57.2 | 1691 | | DeepSeek-R1-Distill-Llama-8B | 50.4 | 80.0 | 89.1 | 49.0 | 39.6 | 1205 | | DeepSeek-R1-Distill-Llama-70B | 70.0 | **86.7** | **94.5** | **65.2** | **57.5** | 1633 | </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` **NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output.** ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).

Sakalti/SJT-4.5B

Sakalti

2025-01-20T23:47:23Z

152

0

transformers

[ "transformers", "safetensors", "qwen2", "text-generation", "text-generation-inference", "unsloth", "trl", "conversational", "en", "ja", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "4-bit", "bitsandbytes", "region:us" ]

text-generation

2024-12-23T08:33:51Z

--- base_model: unsloth/qwen2.5-7b-instruct-bnb-4bit inference: true tags: - text-generation-inference - transformers - unsloth - qwen2 - trl license: apache-2.0 language: - en - ja widget: - messages: - role: user content: こんにちは！ - messages: - role: user content: 魚を捌くのは難しいですか？ - messages: - role: user content: 日本の首都はどこですか？ - messages: - role: user content: hello! - messages: - role: user content: こんにちは！ - messages: - role: user content: whats is the capital of japan? - messages: - role: user content: Who are you? - messages: - role: user content: 你好 --- # Uploaded model - **Developed by:** Sakalti - **License:** apache-2.0 - **Finetuned from model :** unsloth/qwen2.5-7b-instruct-bnb-4bit This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)

JacksonBrune/bf94c46f-78a2-4147-b4e1-1ca46e6cb0ef

JacksonBrune

2025-01-20T23:45:12Z

6

0

peft

[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "base_model:unsloth/Phi-3-medium-4k-instruct", "base_model:adapter:unsloth/Phi-3-medium-4k-instruct", "license:mit", "region:us" ]

null

2025-01-20T22:29:38Z

--- library_name: peft license: mit base_model: unsloth/Phi-3-medium-4k-instruct tags: - axolotl - generated_from_trainer model-index: - name: bf94c46f-78a2-4147-b4e1-1ca46e6cb0ef results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Phi-3-medium-4k-instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 41d403c8b37c92fc_train_data.json ds_type: json format: custom path: /workspace/input_data/41d403c8b37c92fc_train_data.json type: field_input: mesh_terms field_instruction: title field_output: abstract format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: JacksonBrune/bf94c46f-78a2-4147-b4e1-1ca46e6cb0ef hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/41d403c8b37c92fc_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075ea541-bd04-429e-a989-c49dabc36fc3 wandb_project: birthdya-sn56-18-Gradients-On-Demand wandb_run: your_name wandb_runid: 075ea541-bd04-429e-a989-c49dabc36fc3 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # bf94c46f-78a2-4147-b4e1-1ca46e6cb0ef This model is a fine-tuned version of [unsloth/Phi-3-medium-4k-instruct](https://huggingface.co/unsloth/Phi-3-medium-4k-instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.0000 | 1 | nan | | 0.0 | 0.0001 | 3 | nan | | 0.0 | 0.0001 | 6 | nan | | 0.0 | 0.0002 | 9 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

marialvsantiago/7b9a7390-1246-446a-87dd-03f5c20759c1

marialvsantiago

2025-01-20T23:45:06Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:unsloth/SmolLM-1.7B", "base_model:adapter:unsloth/SmolLM-1.7B", "license:apache-2.0", "region:us" ]

null

2025-01-20T23:15:21Z

--- library_name: peft license: apache-2.0 base_model: unsloth/SmolLM-1.7B tags: - axolotl - generated_from_trainer model-index: - name: 7b9a7390-1246-446a-87dd-03f5c20759c1 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/SmolLM-1.7B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - b50bb242ea24ad3f_train_data.json ds_type: json format: custom path: /workspace/input_data/b50bb242ea24ad3f_train_data.json type: field_input: my_solu field_instruction: prompt field_output: solution format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: marialvsantiago/7b9a7390-1246-446a-87dd-03f5c20759c1 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 78GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/b50bb242ea24ad3f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 8caf838c-eaff-45bc-b751-9573db70c518 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 8caf838c-eaff-45bc-b751-9573db70c518 warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # 7b9a7390-1246-446a-87dd-03f5c20759c1 This model is a fine-tuned version of [unsloth/SmolLM-1.7B](https://huggingface.co/unsloth/SmolLM-1.7B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0001 | 1 | nan | | 0.0 | 0.0005 | 5 | nan | | 0.0 | 0.0010 | 10 | nan | | 0.0 | 0.0015 | 15 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

mmnga/DeepSeek-R1-Distill-Qwen-32B-gguf

mmnga

2025-01-20T23:44:54Z

3,380

3

null

[ "gguf", "qwen2", "en", "ja", "dataset:TFMC/imatrix-dataset-for-japanese-llm", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", "license:mit", "endpoints_compatible", "region:us", "imatrix", "conversational" ]

null

2025-01-20T16:54:59Z

--- license: mit language: - en - ja datasets: - TFMC/imatrix-dataset-for-japanese-llm tags: - qwen2 base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --- # DeepSeek-R1-Distill-Qwen-32B-gguf [deepseek-aiさんが公開しているDeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B)のggufフォーマット変換版です。 imatrixのデータは[TFMC/imatrix-dataset-for-japanese-llm](https://huggingface.co/datasets/TFMC/imatrix-dataset-for-japanese-llm)を使用して作成しました。 ## Usage ``` git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build -DGGML_CUDA=ON cmake --build build --config Release build/bin/llama-cli -m 'DeepSeek-R1-Distill-Qwen-32B-gguf' -n 128 -c 128 -p 'あなたはプロの料理人です。レシピを教えて' -cnv ```

Tejasvisudugureddy/distilled_dpo_chatbot

Tejasvisudugureddy

2025-01-20T23:44:50Z

213

0

transformers

[ "transformers", "tensorboard", "safetensors", "llama", "text-generation", "generated_from_trainer", "conversational", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]

text-generation

2025-01-20T23:44:34Z

--- library_name: transformers tags: - generated_from_trainer model-index: - name: distilled_dpo_chatbot results: [] ---  # distilled_dpo_chatbot This model was trained from scratch on an unknown dataset. ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 3.0 ### Framework versions - Transformers 4.47.0 - Pytorch 2.5.1+cu121 - Datasets 3.2.0 - Tokenizers 0.21.0

LHRuig/proultra

LHRuig

2025-01-20T23:44:49Z

9

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:44:29Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # proultra <Gallery /> ## Model description proultra lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/proultra/tree/main) them in the Files & versions tab.

LHRuig/reallora

LHRuig

2025-01-20T23:41:59Z

7

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:41:55Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # reallora <Gallery /> ## Model description reallora lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/reallora/tree/main) them in the Files & versions tab.

LHRuig/db0real

LHRuig

2025-01-20T23:41:19Z

16

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:41:17Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # db0real <Gallery /> ## Model description db0real lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/db0real/tree/main) them in the Files & versions tab.

MayBashendy/ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k19_task5_organization

MayBashendy

2025-01-20T23:41:10Z

5

0

transformers

[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]

text-classification

2025-01-20T23:22:32Z

--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k19_task5_organization results: [] ---  # ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k19_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.3225 - Qwk: 0.0 - Mse: 1.3225 - Rmse: 1.1500 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | Qwk | Mse | Rmse | |:-------------:|:-------:|:----:|:---------------:|:-------:|:------:|:------:| | No log | 0.0435 | 2 | 3.9365 | -0.0232 | 3.9365 | 1.9841 | | No log | 0.0870 | 4 | 2.7213 | -0.0305 | 2.7213 | 1.6496 | | No log | 0.1304 | 6 | 2.3416 | -0.0372 | 2.3416 | 1.5302 | | No log | 0.1739 | 8 | 1.4991 | 0.0854 | 1.4991 | 1.2244 | | No log | 0.2174 | 10 | 1.1113 | 0.2268 | 1.1113 | 1.0542 | | No log | 0.2609 | 12 | 1.1950 | 0.0731 | 1.1950 | 1.0932 | | No log | 0.3043 | 14 | 1.1082 | 0.2094 | 1.1082 | 1.0527 | | No log | 0.3478 | 16 | 1.2264 | 0.0884 | 1.2264 | 1.1074 | | No log | 0.3913 | 18 | 1.6641 | -0.0661 | 1.6641 | 1.2900 | | No log | 0.4348 | 20 | 1.5899 | -0.1122 | 1.5899 | 1.2609 | | No log | 0.4783 | 22 | 1.3669 | 0.0519 | 1.3669 | 1.1692 | | No log | 0.5217 | 24 | 1.3565 | -0.0417 | 1.3565 | 1.1647 | | No log | 0.5652 | 26 | 1.3698 | -0.0032 | 1.3698 | 1.1704 | | No log | 0.6087 | 28 | 1.3109 | -0.0497 | 1.3109 | 1.1450 | | No log | 0.6522 | 30 | 1.0949 | 0.1589 | 1.0949 | 1.0464 | | No log | 0.6957 | 32 | 1.0503 | 0.2114 | 1.0503 | 1.0248 | | No log | 0.7391 | 34 | 1.1274 | 0.1160 | 1.1274 | 1.0618 | | No log | 0.7826 | 36 | 1.4414 | -0.0881 | 1.4414 | 1.2006 | | No log | 0.8261 | 38 | 1.6338 | -0.1445 | 1.6338 | 1.2782 | | No log | 0.8696 | 40 | 1.2646 | 0.0622 | 1.2646 | 1.1245 | | No log | 0.9130 | 42 | 1.0736 | 0.1864 | 1.0736 | 1.0361 | | No log | 0.9565 | 44 | 1.0625 | 0.2911 | 1.0625 | 1.0308 | | No log | 1.0 | 46 | 1.1263 | 0.2647 | 1.1263 | 1.0613 | | No log | 1.0435 | 48 | 1.3547 | -0.0249 | 1.3547 | 1.1639 | | No log | 1.0870 | 50 | 1.6222 | -0.1135 | 1.6222 | 1.2737 | | No log | 1.1304 | 52 | 1.4905 | -0.0623 | 1.4905 | 1.2209 | | No log | 1.1739 | 54 | 1.2550 | 0.0911 | 1.2550 | 1.1203 | | No log | 1.2174 | 56 | 1.0641 | 0.1857 | 1.0641 | 1.0315 | | No log | 1.2609 | 58 | 1.0561 | 0.2179 | 1.0561 | 1.0277 | | No log | 1.3043 | 60 | 1.1071 | 0.1761 | 1.1071 | 1.0522 | | No log | 1.3478 | 62 | 1.2148 | 0.2021 | 1.2148 | 1.1022 | | No log | 1.3913 | 64 | 1.3043 | 0.1489 | 1.3043 | 1.1421 | | No log | 1.4348 | 66 | 1.2530 | 0.1609 | 1.2530 | 1.1194 | | No log | 1.4783 | 68 | 1.1703 | 0.2312 | 1.1703 | 1.0818 | | No log | 1.5217 | 70 | 1.1921 | 0.1801 | 1.1921 | 1.0918 | | No log | 1.5652 | 72 | 1.2234 | 0.1370 | 1.2234 | 1.1061 | | No log | 1.6087 | 74 | 1.4115 | 0.0939 | 1.4115 | 1.1881 | | No log | 1.6522 | 76 | 1.5486 | 0.1111 | 1.5486 | 1.2444 | | No log | 1.6957 | 78 | 1.4303 | 0.0861 | 1.4303 | 1.1959 | | No log | 1.7391 | 80 | 1.2314 | 0.1500 | 1.2314 | 1.1097 | | No log | 1.7826 | 82 | 1.0416 | 0.2226 | 1.0416 | 1.0206 | | No log | 1.8261 | 84 | 0.9985 | 0.2818 | 0.9985 | 0.9993 | | No log | 1.8696 | 86 | 1.0136 | 0.2251 | 1.0136 | 1.0068 | | No log | 1.9130 | 88 | 1.1889 | 0.0961 | 1.1889 | 1.0904 | | No log | 1.9565 | 90 | 1.3523 | 0.0541 | 1.3523 | 1.1629 | | No log | 2.0 | 92 | 1.3806 | 0.0541 | 1.3806 | 1.1750 | | No log | 2.0435 | 94 | 1.2957 | 0.1170 | 1.2957 | 1.1383 | | No log | 2.0870 | 96 | 1.2446 | 0.2577 | 1.2446 | 1.1156 | | No log | 2.1304 | 98 | 1.2931 | 0.2934 | 1.2931 | 1.1371 | | No log | 2.1739 | 100 | 1.3752 | 0.2647 | 1.3752 | 1.1727 | | No log | 2.2174 | 102 | 1.5603 | 0.2239 | 1.5603 | 1.2491 | | No log | 2.2609 | 104 | 1.5673 | 0.1943 | 1.5673 | 1.2519 | | No log | 2.3043 | 106 | 1.3581 | 0.1880 | 1.3581 | 1.1654 | | No log | 2.3478 | 108 | 1.2639 | 0.2203 | 1.2639 | 1.1242 | | No log | 2.3913 | 110 | 1.3721 | 0.1744 | 1.3721 | 1.1713 | | No log | 2.4348 | 112 | 1.3828 | 0.1744 | 1.3828 | 1.1759 | | No log | 2.4783 | 114 | 1.3706 | 0.0931 | 1.3706 | 1.1707 | | No log | 2.5217 | 116 | 1.3150 | 0.0278 | 1.3150 | 1.1467 | | No log | 2.5652 | 118 | 1.3244 | 0.0278 | 1.3244 | 1.1508 | | No log | 2.6087 | 120 | 1.3610 | 0.0571 | 1.3610 | 1.1666 | | No log | 2.6522 | 122 | 1.4379 | 0.1438 | 1.4379 | 1.1991 | | No log | 2.6957 | 124 | 1.5829 | 0.2084 | 1.5829 | 1.2581 | | No log | 2.7391 | 126 | 1.6933 | 0.1630 | 1.6933 | 1.3013 | | No log | 2.7826 | 128 | 1.7564 | 0.1565 | 1.7564 | 1.3253 | | No log | 2.8261 | 130 | 1.6753 | 0.1531 | 1.6753 | 1.2943 | | No log | 2.8696 | 132 | 1.5731 | 0.2084 | 1.5731 | 1.2542 | | No log | 2.9130 | 134 | 1.3438 | 0.1935 | 1.3438 | 1.1592 | | No log | 2.9565 | 136 | 1.1707 | 0.3322 | 1.1707 | 1.0820 | | No log | 3.0 | 138 | 1.1864 | 0.2986 | 1.1864 | 1.0892 | | No log | 3.0435 | 140 | 1.3843 | 0.1727 | 1.3843 | 1.1765 | | No log | 3.0870 | 142 | 1.3959 | 0.2026 | 1.3959 | 1.1815 | | No log | 3.1304 | 144 | 1.2520 | 0.1750 | 1.2520 | 1.1189 | | No log | 3.1739 | 146 | 1.0363 | 0.2836 | 1.0363 | 1.0180 | | No log | 3.2174 | 148 | 1.0039 | 0.2814 | 1.0039 | 1.0020 | | No log | 3.2609 | 150 | 1.0200 | 0.2747 | 1.0200 | 1.0099 | | No log | 3.3043 | 152 | 1.1611 | 0.1751 | 1.1611 | 1.0775 | | No log | 3.3478 | 154 | 1.2712 | 0.2241 | 1.2712 | 1.1275 | | No log | 3.3913 | 156 | 1.3275 | 0.2026 | 1.3275 | 1.1522 | | No log | 3.4348 | 158 | 1.3712 | 0.2015 | 1.3712 | 1.1710 | | No log | 3.4783 | 160 | 1.4292 | 0.2126 | 1.4292 | 1.1955 | | No log | 3.5217 | 162 | 1.3521 | 0.1744 | 1.3521 | 1.1628 | | No log | 3.5652 | 164 | 1.2403 | 0.1052 | 1.2403 | 1.1137 | | No log | 3.6087 | 166 | 1.2213 | 0.1052 | 1.2213 | 1.1051 | | No log | 3.6522 | 168 | 1.2553 | 0.1407 | 1.2553 | 1.1204 | | No log | 3.6957 | 170 | 1.3345 | 0.1744 | 1.3345 | 1.1552 | | No log | 3.7391 | 172 | 1.3148 | 0.1407 | 1.3148 | 1.1467 | | No log | 3.7826 | 174 | 1.2434 | 0.1407 | 1.2434 | 1.1151 | | No log | 3.8261 | 176 | 1.1953 | 0.1744 | 1.1953 | 1.0933 | | No log | 3.8696 | 178 | 1.1782 | 0.1838 | 1.1782 | 1.0855 | | No log | 3.9130 | 180 | 1.2119 | 0.2372 | 1.2119 | 1.1009 | | No log | 3.9565 | 182 | 1.2574 | 0.1952 | 1.2574 | 1.1213 | | No log | 4.0 | 184 | 1.2950 | 0.1288 | 1.2950 | 1.1380 | | No log | 4.0435 | 186 | 1.2514 | 0.1952 | 1.2514 | 1.1187 | | No log | 4.0870 | 188 | 1.0793 | 0.1628 | 1.0793 | 1.0389 | | No log | 4.1304 | 190 | 0.9864 | 0.0990 | 0.9864 | 0.9932 | | No log | 4.1739 | 192 | 1.0281 | 0.1654 | 1.0281 | 1.0139 | | No log | 4.2174 | 194 | 1.2053 | 0.1943 | 1.2053 | 1.0979 | | No log | 4.2609 | 196 | 1.4822 | 0.2733 | 1.4822 | 1.2174 | | No log | 4.3043 | 198 | 1.6190 | 0.2733 | 1.6190 | 1.2724 | | No log | 4.3478 | 200 | 1.5802 | 0.2568 | 1.5802 | 1.2571 | | No log | 4.3913 | 202 | 1.4305 | 0.2089 | 1.4305 | 1.1961 | | No log | 4.4348 | 204 | 1.2446 | 0.1654 | 1.2446 | 1.1156 | | No log | 4.4783 | 206 | 1.1579 | 0.1226 | 1.1579 | 1.0761 | | No log | 4.5217 | 208 | 1.2399 | -0.0929 | 1.2399 | 1.1135 | | No log | 4.5652 | 210 | 1.3674 | -0.0263 | 1.3674 | 1.1693 | | No log | 4.6087 | 212 | 1.4571 | 0.0935 | 1.4571 | 1.2071 | | No log | 4.6522 | 214 | 1.6547 | 0.1565 | 1.6547 | 1.2864 | | No log | 4.6957 | 216 | 1.7420 | 0.1476 | 1.7420 | 1.3198 | | No log | 4.7391 | 218 | 1.6654 | 0.1423 | 1.6654 | 1.2905 | | No log | 4.7826 | 220 | 1.4261 | 0.2062 | 1.4261 | 1.1942 | | No log | 4.8261 | 222 | 1.1989 | 0.2038 | 1.1989 | 1.0950 | | No log | 4.8696 | 224 | 1.1364 | 0.1976 | 1.1364 | 1.0660 | | No log | 4.9130 | 226 | 1.1866 | 0.2126 | 1.1866 | 1.0893 | | No log | 4.9565 | 228 | 1.3037 | 0.2126 | 1.3037 | 1.1418 | | No log | 5.0 | 230 | 1.4564 | 0.2342 | 1.4564 | 1.2068 | | No log | 5.0435 | 232 | 1.5256 | 0.2694 | 1.5256 | 1.2352 | | No log | 5.0870 | 234 | 1.5762 | 0.2940 | 1.5762 | 1.2555 | | No log | 5.1304 | 236 | 1.5906 | 0.2940 | 1.5906 | 1.2612 | | No log | 5.1739 | 238 | 1.6182 | 0.2974 | 1.6182 | 1.2721 | | No log | 5.2174 | 240 | 1.6432 | 0.2974 | 1.6432 | 1.2819 | | No log | 5.2609 | 242 | 1.5135 | 0.2694 | 1.5135 | 1.2302 | | No log | 5.3043 | 244 | 1.3818 | 0.2170 | 1.3818 | 1.1755 | | No log | 5.3478 | 246 | 1.2951 | 0.2123 | 1.2951 | 1.1380 | | No log | 5.3913 | 248 | 1.2601 | 0.1850 | 1.2601 | 1.1225 | | No log | 5.4348 | 250 | 1.2575 | 0.2170 | 1.2574 | 1.1214 | | No log | 5.4783 | 252 | 1.2783 | 0.2117 | 1.2783 | 1.1306 | | No log | 5.5217 | 254 | 1.3535 | 0.2117 | 1.3535 | 1.1634 | | No log | 5.5652 | 256 | 1.4060 | 0.2391 | 1.4060 | 1.1857 | | No log | 5.6087 | 258 | 1.3457 | 0.2117 | 1.3457 | 1.1600 | | No log | 5.6522 | 260 | 1.2294 | 0.2170 | 1.2294 | 1.1088 | | No log | 5.6957 | 262 | 1.2760 | 0.2170 | 1.2760 | 1.1296 | | No log | 5.7391 | 264 | 1.4787 | 0.2653 | 1.4787 | 1.2160 | | No log | 5.7826 | 266 | 1.5640 | 0.2391 | 1.5640 | 1.2506 | | No log | 5.8261 | 268 | 1.5706 | 0.2062 | 1.5706 | 1.2532 | | No log | 5.8696 | 270 | 1.4463 | 0.2239 | 1.4463 | 1.2026 | | No log | 5.9130 | 272 | 1.2451 | 0.1407 | 1.2451 | 1.1158 | | No log | 5.9565 | 274 | 1.1629 | 0.0401 | 1.1629 | 1.0784 | | No log | 6.0 | 276 | 1.1448 | 0.0401 | 1.1448 | 1.0699 | | No log | 6.0435 | 278 | 1.1363 | 0.0 | 1.1363 | 1.0660 | | No log | 6.0870 | 280 | 1.2335 | 0.0781 | 1.2335 | 1.1106 | | No log | 6.1304 | 282 | 1.4538 | 0.1814 | 1.4538 | 1.2057 | | No log | 6.1739 | 284 | 1.6059 | 0.2694 | 1.6059 | 1.2673 | | No log | 6.2174 | 286 | 1.5940 | 0.2315 | 1.5940 | 1.2625 | | No log | 6.2609 | 288 | 1.5500 | 0.2270 | 1.5500 | 1.2450 | | No log | 6.3043 | 290 | 1.4412 | 0.1850 | 1.4412 | 1.2005 | | No log | 6.3478 | 292 | 1.3873 | 0.1769 | 1.3873 | 1.1778 | | No log | 6.3913 | 294 | 1.2601 | 0.0 | 1.2601 | 1.1226 | | No log | 6.4348 | 296 | 1.2080 | 0.0 | 1.2080 | 1.0991 | | No log | 6.4783 | 298 | 1.2249 | 0.0 | 1.2249 | 1.1067 | | No log | 6.5217 | 300 | 1.2305 | 0.0 | 1.2305 | 1.1093 | | No log | 6.5652 | 302 | 1.2659 | 0.0033 | 1.2659 | 1.1251 | | No log | 6.6087 | 304 | 1.3705 | 0.1255 | 1.3705 | 1.1707 | | No log | 6.6522 | 306 | 1.5111 | 0.1370 | 1.5111 | 1.2293 | | No log | 6.6957 | 308 | 1.6192 | 0.2342 | 1.6192 | 1.2725 | | No log | 6.7391 | 310 | 1.5472 | 0.2424 | 1.5472 | 1.2439 | | No log | 6.7826 | 312 | 1.4260 | 0.1113 | 1.4260 | 1.1942 | | No log | 6.8261 | 314 | 1.3601 | 0.0 | 1.3601 | 1.1662 | | No log | 6.8696 | 316 | 1.4354 | 0.0 | 1.4354 | 1.1981 | | No log | 6.9130 | 318 | 1.4593 | 0.0 | 1.4593 | 1.2080 | | No log | 6.9565 | 320 | 1.5025 | 0.0661 | 1.5025 | 1.2258 | | No log | 7.0 | 322 | 1.5379 | 0.1052 | 1.5379 | 1.2401 | | No log | 7.0435 | 324 | 1.5436 | 0.1744 | 1.5436 | 1.2424 | | No log | 7.0870 | 326 | 1.5589 | 0.0661 | 1.5589 | 1.2486 | | No log | 7.1304 | 328 | 1.5395 | 0.0883 | 1.5395 | 1.2408 | | No log | 7.1739 | 330 | 1.5589 | 0.1734 | 1.5589 | 1.2485 | | No log | 7.2174 | 332 | 1.7114 | 0.1808 | 1.7114 | 1.3082 | | No log | 7.2609 | 334 | 1.8738 | 0.1002 | 1.8738 | 1.3689 | | No log | 7.3043 | 336 | 1.9387 | 0.1002 | 1.9387 | 1.3924 | | No log | 7.3478 | 338 | 1.8955 | 0.1807 | 1.8955 | 1.3768 | | No log | 7.3913 | 340 | 1.6579 | 0.2117 | 1.6579 | 1.2876 | | No log | 7.4348 | 342 | 1.4418 | 0.1288 | 1.4418 | 1.2008 | | No log | 7.4783 | 344 | 1.3060 | 0.0931 | 1.3060 | 1.1428 | | No log | 7.5217 | 346 | 1.3268 | 0.0541 | 1.3268 | 1.1519 | | No log | 7.5652 | 348 | 1.3956 | 0.1744 | 1.3956 | 1.1813 | | No log | 7.6087 | 350 | 1.5431 | 0.2474 | 1.5431 | 1.2422 | | No log | 7.6522 | 352 | 1.6069 | 0.2611 | 1.6069 | 1.2676 | | No log | 7.6957 | 354 | 1.6304 | 0.2174 | 1.6304 | 1.2769 | | No log | 7.7391 | 356 | 1.5382 | 0.2123 | 1.5382 | 1.2402 | | No log | 7.7826 | 358 | 1.3799 | 0.1727 | 1.3799 | 1.1747 | | No log | 7.8261 | 360 | 1.2919 | 0.1316 | 1.2919 | 1.1366 | | No log | 7.8696 | 362 | 1.3619 | 0.1024 | 1.3619 | 1.1670 | | No log | 7.9130 | 364 | 1.4500 | 0.2239 | 1.4500 | 1.2041 | | No log | 7.9565 | 366 | 1.4819 | 0.2342 | 1.4819 | 1.2173 | | No log | 8.0 | 368 | 1.3776 | 0.1197 | 1.3776 | 1.1737 | | No log | 8.0435 | 370 | 1.1953 | 0.1019 | 1.1953 | 1.0933 | | No log | 8.0870 | 372 | 1.1063 | 0.0406 | 1.1063 | 1.0518 | | No log | 8.1304 | 374 | 1.1168 | 0.0406 | 1.1168 | 1.0568 | | No log | 8.1739 | 376 | 1.2230 | 0.1705 | 1.2230 | 1.1059 | | No log | 8.2174 | 378 | 1.4200 | 0.1370 | 1.4200 | 1.1916 | | No log | 8.2609 | 380 | 1.4560 | 0.0931 | 1.4560 | 1.2066 | | No log | 8.3043 | 382 | 1.4686 | 0.0931 | 1.4686 | 1.2119 | | No log | 8.3478 | 384 | 1.4976 | 0.1744 | 1.4976 | 1.2237 | | No log | 8.3913 | 386 | 1.4849 | 0.1744 | 1.4849 | 1.2186 | | No log | 8.4348 | 388 | 1.5288 | 0.2665 | 1.5288 | 1.2365 | | No log | 8.4783 | 390 | 1.5550 | 0.2752 | 1.5550 | 1.2470 | | No log | 8.5217 | 392 | 1.4539 | 0.2372 | 1.4539 | 1.2058 | | No log | 8.5652 | 394 | 1.3677 | 0.2065 | 1.3677 | 1.1695 | | No log | 8.6087 | 396 | 1.3685 | 0.1407 | 1.3685 | 1.1698 | | No log | 8.6522 | 398 | 1.4109 | 0.1744 | 1.4109 | 1.1878 | | No log | 8.6957 | 400 | 1.5116 | 0.2424 | 1.5116 | 1.2295 | | No log | 8.7391 | 402 | 1.5118 | 0.2474 | 1.5118 | 1.2295 | | No log | 8.7826 | 404 | 1.5575 | 0.2522 | 1.5575 | 1.2480 | | No log | 8.8261 | 406 | 1.5527 | 0.2126 | 1.5527 | 1.2461 | | No log | 8.8696 | 408 | 1.4464 | 0.1407 | 1.4464 | 1.2027 | | No log | 8.9130 | 410 | 1.4063 | 0.0 | 1.4063 | 1.1859 | | No log | 8.9565 | 412 | 1.3768 | 0.0 | 1.3768 | 1.1734 | | No log | 9.0 | 414 | 1.3550 | -0.0411 | 1.3550 | 1.1641 | | No log | 9.0435 | 416 | 1.3643 | 0.0 | 1.3643 | 1.1680 | | No log | 9.0870 | 418 | 1.4810 | 0.0401 | 1.4810 | 1.2170 | | No log | 9.1304 | 420 | 1.6359 | 0.2126 | 1.6359 | 1.2790 | | No log | 9.1739 | 422 | 1.7412 | 0.1892 | 1.7412 | 1.3196 | | No log | 9.2174 | 424 | 1.8050 | 0.1487 | 1.8050 | 1.3435 | | No log | 9.2609 | 426 | 1.7355 | 0.1955 | 1.7355 | 1.3174 | | No log | 9.3043 | 428 | 1.5760 | 0.2653 | 1.5760 | 1.2554 | | No log | 9.3478 | 430 | 1.4390 | 0.2015 | 1.4390 | 1.1996 | | No log | 9.3913 | 432 | 1.3894 | 0.1700 | 1.3894 | 1.1787 | | No log | 9.4348 | 434 | 1.3953 | 0.1744 | 1.3953 | 1.1812 | | No log | 9.4783 | 436 | 1.4498 | 0.1744 | 1.4498 | 1.2041 | | No log | 9.5217 | 438 | 1.4705 | 0.1744 | 1.4705 | 1.2126 | | No log | 9.5652 | 440 | 1.3951 | 0.1744 | 1.3951 | 1.1811 | | No log | 9.6087 | 442 | 1.2963 | 0.1052 | 1.2963 | 1.1385 | | No log | 9.6522 | 444 | 1.2880 | 0.1052 | 1.2880 | 1.1349 | | No log | 9.6957 | 446 | 1.3230 | 0.1407 | 1.3230 | 1.1502 | | No log | 9.7391 | 448 | 1.3931 | 0.1407 | 1.3931 | 1.1803 | | No log | 9.7826 | 450 | 1.4239 | 0.1052 | 1.4239 | 1.1933 | | No log | 9.8261 | 452 | 1.4655 | 0.1744 | 1.4655 | 1.2106 | | No log | 9.8696 | 454 | 1.4982 | 0.2372 | 1.4982 | 1.2240 | | No log | 9.9130 | 456 | 1.5002 | 0.2372 | 1.5002 | 1.2248 | | No log | 9.9565 | 458 | 1.4132 | 0.2372 | 1.4132 | 1.1888 | | No log | 10.0 | 460 | 1.2400 | 0.0 | 1.2400 | 1.1135 | | No log | 10.0435 | 462 | 1.1297 | 0.0155 | 1.1297 | 1.0629 | | No log | 10.0870 | 464 | 1.1569 | 0.0155 | 1.1569 | 1.0756 | | No log | 10.1304 | 466 | 1.2550 | 0.0 | 1.2550 | 1.1203 | | No log | 10.1739 | 468 | 1.2879 | 0.1113 | 1.2879 | 1.1348 | | No log | 10.2174 | 470 | 1.3146 | 0.0390 | 1.3146 | 1.1466 | | No log | 10.2609 | 472 | 1.3950 | 0.0781 | 1.3950 | 1.1811 | | No log | 10.3043 | 474 | 1.4311 | 0.0781 | 1.4311 | 1.1963 | | No log | 10.3478 | 476 | 1.4025 | 0.0781 | 1.4025 | 1.1843 | | No log | 10.3913 | 478 | 1.3441 | 0.0 | 1.3441 | 1.1593 | | No log | 10.4348 | 480 | 1.3662 | 0.0 | 1.3662 | 1.1689 | | No log | 10.4783 | 482 | 1.4121 | 0.1370 | 1.4121 | 1.1883 | | No log | 10.5217 | 484 | 1.5203 | 0.2015 | 1.5203 | 1.2330 | | No log | 10.5652 | 486 | 1.7276 | 0.1963 | 1.7276 | 1.3144 | | No log | 10.6087 | 488 | 1.8236 | 0.2056 | 1.8236 | 1.3504 | | No log | 10.6522 | 490 | 1.7907 | 0.2317 | 1.7907 | 1.3382 | | No log | 10.6957 | 492 | 1.6974 | 0.2391 | 1.6974 | 1.3028 | | No log | 10.7391 | 494 | 1.5156 | 0.1744 | 1.5156 | 1.2311 | | No log | 10.7826 | 496 | 1.4285 | 0.0390 | 1.4285 | 1.1952 | | No log | 10.8261 | 498 | 1.3874 | 0.0390 | 1.3874 | 1.1779 | | 0.2613 | 10.8696 | 500 | 1.3506 | 0.0 | 1.3506 | 1.1622 | | 0.2613 | 10.9130 | 502 | 1.3637 | 0.0661 | 1.3637 | 1.1678 | | 0.2613 | 10.9565 | 504 | 1.4279 | 0.1370 | 1.4279 | 1.1950 | | 0.2613 | 11.0 | 506 | 1.5500 | 0.2315 | 1.5500 | 1.2450 | | 0.2613 | 11.0435 | 508 | 1.6369 | 0.2465 | 1.6369 | 1.2794 | | 0.2613 | 11.0870 | 510 | 1.7018 | 0.2832 | 1.7018 | 1.3045 | | 0.2613 | 11.1304 | 512 | 1.6085 | 0.2665 | 1.6085 | 1.2683 | | 0.2613 | 11.1739 | 514 | 1.4327 | 0.1744 | 1.4327 | 1.1969 | | 0.2613 | 11.2174 | 516 | 1.3146 | 0.0401 | 1.3146 | 1.1466 | | 0.2613 | 11.2609 | 518 | 1.2902 | 0.0401 | 1.2902 | 1.1359 | | 0.2613 | 11.3043 | 520 | 1.3147 | 0.0401 | 1.3147 | 1.1466 | | 0.2613 | 11.3478 | 522 | 1.3176 | 0.1407 | 1.3176 | 1.1479 | | 0.2613 | 11.3913 | 524 | 1.3261 | 0.1744 | 1.3261 | 1.1516 | | 0.2613 | 11.4348 | 526 | 1.2363 | 0.2206 | 1.2363 | 1.1119 | | 0.2613 | 11.4783 | 528 | 1.2522 | 0.2313 | 1.2522 | 1.1190 | | 0.2613 | 11.5217 | 530 | 1.4130 | 0.2431 | 1.4130 | 1.1887 | | 0.2613 | 11.5652 | 532 | 1.7543 | 0.1911 | 1.7543 | 1.3245 | | 0.2613 | 11.6087 | 534 | 1.8516 | 0.2406 | 1.8516 | 1.3607 | | 0.2613 | 11.6522 | 536 | 1.7143 | 0.1892 | 1.7143 | 1.3093 | | 0.2613 | 11.6957 | 538 | 1.5454 | 0.1288 | 1.5454 | 1.2431 | | 0.2613 | 11.7391 | 540 | 1.4371 | 0.0781 | 1.4371 | 1.1988 | | 0.2613 | 11.7826 | 542 | 1.3778 | 0.1370 | 1.3778 | 1.1738 | | 0.2613 | 11.8261 | 544 | 1.3337 | 0.1370 | 1.3337 | 1.1549 | | 0.2613 | 11.8696 | 546 | 1.3540 | 0.1700 | 1.3540 | 1.1636 | | 0.2613 | 11.9130 | 548 | 1.4217 | 0.2126 | 1.4217 | 1.1924 | | 0.2613 | 11.9565 | 550 | 1.3881 | 0.2065 | 1.3881 | 1.1782 | | 0.2613 | 12.0 | 552 | 1.2795 | 0.0781 | 1.2795 | 1.1312 | | 0.2613 | 12.0435 | 554 | 1.1736 | 0.0155 | 1.1736 | 1.0834 | | 0.2613 | 12.0870 | 556 | 1.1141 | 0.0587 | 1.1141 | 1.0555 | | 0.2613 | 12.1304 | 558 | 1.0583 | 0.0618 | 1.0583 | 1.0288 | | 0.2613 | 12.1739 | 560 | 1.0693 | 0.1407 | 1.0693 | 1.0341 | | 0.2613 | 12.2174 | 562 | 1.1314 | 0.0961 | 1.1314 | 1.0637 | | 0.2613 | 12.2609 | 564 | 1.2606 | 0.1769 | 1.2606 | 1.1228 | | 0.2613 | 12.3043 | 566 | 1.3739 | 0.2126 | 1.3739 | 1.1721 | | 0.2613 | 12.3478 | 568 | 1.4466 | 0.2126 | 1.4466 | 1.2028 | | 0.2613 | 12.3913 | 570 | 1.4543 | 0.2126 | 1.4543 | 1.2060 | | 0.2613 | 12.4348 | 572 | 1.5068 | 0.2424 | 1.5068 | 1.2275 | | 0.2613 | 12.4783 | 574 | 1.6092 | 0.2832 | 1.6092 | 1.2685 | | 0.2613 | 12.5217 | 576 | 1.7369 | 0.2940 | 1.7369 | 1.3179 | | 0.2613 | 12.5652 | 578 | 1.7533 | 0.2940 | 1.7533 | 1.3241 | | 0.2613 | 12.6087 | 580 | 1.6560 | 0.2474 | 1.6560 | 1.2868 | | 0.2613 | 12.6522 | 582 | 1.5699 | 0.1228 | 1.5699 | 1.2530 | | 0.2613 | 12.6957 | 584 | 1.5091 | 0.0781 | 1.5091 | 1.2285 | | 0.2613 | 12.7391 | 586 | 1.4348 | 0.0401 | 1.4348 | 1.1978 | | 0.2613 | 12.7826 | 588 | 1.3822 | 0.0401 | 1.3822 | 1.1757 | | 0.2613 | 12.8261 | 590 | 1.3843 | 0.0401 | 1.3843 | 1.1766 | | 0.2613 | 12.8696 | 592 | 1.4302 | 0.1407 | 1.4302 | 1.1959 | | 0.2613 | 12.9130 | 594 | 1.4860 | 0.2424 | 1.4860 | 1.2190 | | 0.2613 | 12.9565 | 596 | 1.5159 | 0.2474 | 1.5159 | 1.2312 | | 0.2613 | 13.0 | 598 | 1.5296 | 0.2793 | 1.5296 | 1.2368 | | 0.2613 | 13.0435 | 600 | 1.6018 | 0.2832 | 1.6018 | 1.2656 | | 0.2613 | 13.0870 | 602 | 1.5336 | 0.2832 | 1.5336 | 1.2384 | | 0.2613 | 13.1304 | 604 | 1.4550 | 0.2474 | 1.4550 | 1.2062 | | 0.2613 | 13.1739 | 606 | 1.4058 | 0.2424 | 1.4058 | 1.1856 | | 0.2613 | 13.2174 | 608 | 1.3353 | 0.2075 | 1.3353 | 1.1556 | | 0.2613 | 13.2609 | 610 | 1.3624 | 0.2075 | 1.3624 | 1.1672 | | 0.2613 | 13.3043 | 612 | 1.3326 | 0.1769 | 1.3326 | 1.1544 | | 0.2613 | 13.3478 | 614 | 1.3241 | 0.1769 | 1.3241 | 1.1507 | | 0.2613 | 13.3913 | 616 | 1.3316 | 0.1769 | 1.3316 | 1.1539 | | 0.2613 | 13.4348 | 618 | 1.3209 | 0.1769 | 1.3209 | 1.1493 | | 0.2613 | 13.4783 | 620 | 1.3151 | 0.1769 | 1.3151 | 1.1468 | | 0.2613 | 13.5217 | 622 | 1.3106 | 0.1769 | 1.3106 | 1.1448 | | 0.2613 | 13.5652 | 624 | 1.3274 | 0.2424 | 1.3274 | 1.1521 | | 0.2613 | 13.6087 | 626 | 1.4479 | 0.2752 | 1.4479 | 1.2033 | | 0.2613 | 13.6522 | 628 | 1.5377 | 0.2568 | 1.5377 | 1.2400 | | 0.2613 | 13.6957 | 630 | 1.5216 | 0.2568 | 1.5216 | 1.2336 | | 0.2613 | 13.7391 | 632 | 1.5517 | 0.2568 | 1.5517 | 1.2457 | | 0.2613 | 13.7826 | 634 | 1.6063 | 0.2568 | 1.6063 | 1.2674 | | 0.2613 | 13.8261 | 636 | 1.5487 | 0.2752 | 1.5487 | 1.2445 | | 0.2613 | 13.8696 | 638 | 1.4303 | 0.2065 | 1.4303 | 1.1960 | | 0.2613 | 13.9130 | 640 | 1.3684 | 0.2065 | 1.3684 | 1.1698 | | 0.2613 | 13.9565 | 642 | 1.3211 | 0.0401 | 1.3211 | 1.1494 | | 0.2613 | 14.0 | 644 | 1.3409 | 0.0401 | 1.3409 | 1.1580 | | 0.2613 | 14.0435 | 646 | 1.4071 | 0.0781 | 1.4071 | 1.1862 | | 0.2613 | 14.0870 | 648 | 1.4992 | 0.2065 | 1.4992 | 1.2244 | | 0.2613 | 14.1304 | 650 | 1.6000 | 0.2709 | 1.6000 | 1.2649 | | 0.2613 | 14.1739 | 652 | 1.6450 | 0.2752 | 1.6450 | 1.2826 | | 0.2613 | 14.2174 | 654 | 1.6421 | 0.2752 | 1.6421 | 1.2814 | | 0.2613 | 14.2609 | 656 | 1.6102 | 0.2417 | 1.6102 | 1.2689 | | 0.2613 | 14.3043 | 658 | 1.5413 | 0.2372 | 1.5413 | 1.2415 | | 0.2613 | 14.3478 | 660 | 1.5115 | 0.2372 | 1.5115 | 1.2294 | | 0.2613 | 14.3913 | 662 | 1.5279 | 0.2372 | 1.5279 | 1.2361 | | 0.2613 | 14.4348 | 664 | 1.5128 | 0.2372 | 1.5128 | 1.2300 | | 0.2613 | 14.4783 | 666 | 1.5337 | 0.2424 | 1.5337 | 1.2384 | | 0.2613 | 14.5217 | 668 | 1.5367 | 0.2752 | 1.5367 | 1.2397 | | 0.2613 | 14.5652 | 670 | 1.5846 | 0.2752 | 1.5846 | 1.2588 | | 0.2613 | 14.6087 | 672 | 1.6122 | 0.2832 | 1.6122 | 1.2697 | | 0.2613 | 14.6522 | 674 | 1.5983 | 0.2752 | 1.5983 | 1.2642 | | 0.2613 | 14.6957 | 676 | 1.5849 | 0.2752 | 1.5849 | 1.2589 | | 0.2613 | 14.7391 | 678 | 1.5418 | 0.2752 | 1.5418 | 1.2417 | | 0.2613 | 14.7826 | 680 | 1.4476 | 0.1744 | 1.4476 | 1.2032 | | 0.2613 | 14.8261 | 682 | 1.3866 | 0.1142 | 1.3866 | 1.1776 | | 0.2613 | 14.8696 | 684 | 1.3755 | 0.0401 | 1.3755 | 1.1728 | | 0.2613 | 14.9130 | 686 | 1.3934 | 0.0781 | 1.3934 | 1.1804 | | 0.2613 | 14.9565 | 688 | 1.4456 | 0.1142 | 1.4456 | 1.2023 | | 0.2613 | 15.0 | 690 | 1.5034 | 0.1769 | 1.5034 | 1.2261 | | 0.2613 | 15.0435 | 692 | 1.5961 | 0.2752 | 1.5961 | 1.2634 | | 0.2613 | 15.0870 | 694 | 1.7191 | 0.2832 | 1.7191 | 1.3111 | | 0.2613 | 15.1304 | 696 | 1.8074 | 0.2406 | 1.8074 | 1.3444 | | 0.2613 | 15.1739 | 698 | 1.7995 | 0.2770 | 1.7995 | 1.3414 | | 0.2613 | 15.2174 | 700 | 1.6816 | 0.2752 | 1.6816 | 1.2968 | | 0.2613 | 15.2609 | 702 | 1.5217 | 0.1142 | 1.5217 | 1.2336 | | 0.2613 | 15.3043 | 704 | 1.3637 | 0.0401 | 1.3637 | 1.1678 | | 0.2613 | 15.3478 | 706 | 1.2650 | 0.0 | 1.2650 | 1.1247 | | 0.2613 | 15.3913 | 708 | 1.2534 | 0.0 | 1.2534 | 1.1196 | | 0.2613 | 15.4348 | 710 | 1.2924 | 0.0 | 1.2924 | 1.1368 | | 0.2613 | 15.4783 | 712 | 1.3811 | 0.0401 | 1.3811 | 1.1752 | | 0.2613 | 15.5217 | 714 | 1.5260 | 0.0781 | 1.5260 | 1.2353 | | 0.2613 | 15.5652 | 716 | 1.6227 | 0.2065 | 1.6227 | 1.2738 | | 0.2613 | 15.6087 | 718 | 1.6985 | 0.2709 | 1.6985 | 1.3033 | | 0.2613 | 15.6522 | 720 | 1.7279 | 0.2709 | 1.7279 | 1.3145 | | 0.2613 | 15.6957 | 722 | 1.6787 | 0.0781 | 1.6787 | 1.2956 | | 0.2613 | 15.7391 | 724 | 1.5888 | 0.0401 | 1.5888 | 1.2605 | | 0.2613 | 15.7826 | 726 | 1.5083 | 0.0 | 1.5083 | 1.2281 | | 0.2613 | 15.8261 | 728 | 1.4668 | 0.0 | 1.4668 | 1.2111 | | 0.2613 | 15.8696 | 730 | 1.4395 | 0.0 | 1.4395 | 1.1998 | | 0.2613 | 15.9130 | 732 | 1.4284 | 0.0401 | 1.4284 | 1.1952 | | 0.2613 | 15.9565 | 734 | 1.4698 | 0.0401 | 1.4698 | 1.2123 | | 0.2613 | 16.0 | 736 | 1.4808 | 0.0781 | 1.4808 | 1.2169 | | 0.2613 | 16.0435 | 738 | 1.4348 | 0.0401 | 1.4348 | 1.1978 | | 0.2613 | 16.0870 | 740 | 1.3826 | 0.0401 | 1.3826 | 1.1758 | | 0.2613 | 16.1304 | 742 | 1.3705 | 0.0401 | 1.3705 | 1.1707 | | 0.2613 | 16.1739 | 744 | 1.4221 | 0.0401 | 1.4221 | 1.1925 | | 0.2613 | 16.2174 | 746 | 1.5149 | 0.0781 | 1.5149 | 1.2308 | | 0.2613 | 16.2609 | 748 | 1.6344 | 0.1142 | 1.6344 | 1.2785 | | 0.2613 | 16.3043 | 750 | 1.6958 | 0.1142 | 1.6958 | 1.3022 | | 0.2613 | 16.3478 | 752 | 1.7174 | 0.1744 | 1.7174 | 1.3105 | | 0.2613 | 16.3913 | 754 | 1.6890 | 0.1744 | 1.6890 | 1.2996 | | 0.2613 | 16.4348 | 756 | 1.6066 | 0.1142 | 1.6066 | 1.2675 | | 0.2613 | 16.4783 | 758 | 1.5513 | 0.0390 | 1.5513 | 1.2455 | | 0.2613 | 16.5217 | 760 | 1.5064 | 0.0390 | 1.5064 | 1.2273 | | 0.2613 | 16.5652 | 762 | 1.4922 | 0.0390 | 1.4922 | 1.2215 | | 0.2613 | 16.6087 | 764 | 1.4445 | 0.0390 | 1.4445 | 1.2019 | | 0.2613 | 16.6522 | 766 | 1.4327 | 0.1407 | 1.4327 | 1.1970 | | 0.2613 | 16.6957 | 768 | 1.4487 | 0.2065 | 1.4487 | 1.2036 | | 0.2613 | 16.7391 | 770 | 1.3882 | 0.0781 | 1.3882 | 1.1782 | | 0.2613 | 16.7826 | 772 | 1.2968 | 0.0781 | 1.2968 | 1.1388 | | 0.2613 | 16.8261 | 774 | 1.2913 | 0.0781 | 1.2913 | 1.1363 | | 0.2613 | 16.8696 | 776 | 1.3576 | 0.2372 | 1.3576 | 1.1652 | | 0.2613 | 16.9130 | 778 | 1.4646 | 0.2424 | 1.4646 | 1.2102 | | 0.2613 | 16.9565 | 780 | 1.4935 | 0.2522 | 1.4935 | 1.2221 | | 0.2613 | 17.0 | 782 | 1.4955 | 0.2709 | 1.4955 | 1.2229 | | 0.2613 | 17.0435 | 784 | 1.4958 | 0.2075 | 1.4958 | 1.2230 | | 0.2613 | 17.0870 | 786 | 1.4286 | 0.1113 | 1.4286 | 1.1952 | | 0.2613 | 17.1304 | 788 | 1.3892 | 0.0155 | 1.3892 | 1.1786 | | 0.2613 | 17.1739 | 790 | 1.3827 | 0.0155 | 1.3827 | 1.1759 | | 0.2613 | 17.2174 | 792 | 1.3759 | 0.0155 | 1.3759 | 1.1730 | | 0.2613 | 17.2609 | 794 | 1.4105 | 0.0155 | 1.4105 | 1.1876 | | 0.2613 | 17.3043 | 796 | 1.4572 | 0.1113 | 1.4572 | 1.2071 | | 0.2613 | 17.3478 | 798 | 1.4896 | 0.2424 | 1.4896 | 1.2205 | | 0.2613 | 17.3913 | 800 | 1.4728 | 0.2424 | 1.4728 | 1.2136 | | 0.2613 | 17.4348 | 802 | 1.4114 | 0.2126 | 1.4114 | 1.1880 | | 0.2613 | 17.4783 | 804 | 1.3548 | 0.0781 | 1.3548 | 1.1640 | | 0.2613 | 17.5217 | 806 | 1.3085 | 0.0401 | 1.3085 | 1.1439 | | 0.2613 | 17.5652 | 808 | 1.2499 | 0.0 | 1.2499 | 1.1180 | | 0.2613 | 17.6087 | 810 | 1.2468 | 0.0 | 1.2468 | 1.1166 | | 0.2613 | 17.6522 | 812 | 1.3334 | 0.0401 | 1.3334 | 1.1547 | | 0.2613 | 17.6957 | 814 | 1.4046 | 0.2424 | 1.4046 | 1.1852 | | 0.2613 | 17.7391 | 816 | 1.4401 | 0.2424 | 1.4401 | 1.2000 | | 0.2613 | 17.7826 | 818 | 1.4071 | 0.2075 | 1.4071 | 1.1862 | | 0.2613 | 17.8261 | 820 | 1.4065 | 0.2367 | 1.4065 | 1.1860 | | 0.2613 | 17.8696 | 822 | 1.3824 | 0.2367 | 1.3824 | 1.1758 | | 0.2613 | 17.9130 | 824 | 1.3945 | 0.2367 | 1.3945 | 1.1809 | | 0.2613 | 17.9565 | 826 | 1.4377 | 0.2647 | 1.4377 | 1.1990 | | 0.2613 | 18.0 | 828 | 1.4406 | 0.2015 | 1.4406 | 1.2002 | | 0.2613 | 18.0435 | 830 | 1.4659 | 0.1370 | 1.4659 | 1.2108 | | 0.2613 | 18.0870 | 832 | 1.4324 | 0.1370 | 1.4324 | 1.1968 | | 0.2613 | 18.1304 | 834 | 1.3917 | 0.0390 | 1.3917 | 1.1797 | | 0.2613 | 18.1739 | 836 | 1.3690 | 0.1024 | 1.3690 | 1.1700 | | 0.2613 | 18.2174 | 838 | 1.3763 | 0.1370 | 1.3763 | 1.1732 | | 0.2613 | 18.2609 | 840 | 1.4114 | 0.1449 | 1.4114 | 1.1880 | | 0.2613 | 18.3043 | 842 | 1.4059 | 0.1769 | 1.4059 | 1.1857 | | 0.2613 | 18.3478 | 844 | 1.4071 | 0.2038 | 1.4071 | 1.1862 | | 0.2613 | 18.3913 | 846 | 1.4425 | 0.1663 | 1.4425 | 1.2010 | | 0.2613 | 18.4348 | 848 | 1.5222 | 0.2291 | 1.5222 | 1.2338 | | 0.2613 | 18.4783 | 850 | 1.4608 | 0.1769 | 1.4608 | 1.2086 | | 0.2613 | 18.5217 | 852 | 1.3935 | 0.0390 | 1.3935 | 1.1805 | | 0.2613 | 18.5652 | 854 | 1.3725 | 0.0 | 1.3725 | 1.1715 | | 0.2613 | 18.6087 | 856 | 1.3666 | 0.0390 | 1.3666 | 1.1690 | | 0.2613 | 18.6522 | 858 | 1.3549 | 0.0760 | 1.3549 | 1.1640 | | 0.2613 | 18.6957 | 860 | 1.3186 | 0.0760 | 1.3186 | 1.1483 | | 0.2613 | 18.7391 | 862 | 1.2866 | 0.0760 | 1.2866 | 1.1343 | | 0.2613 | 18.7826 | 864 | 1.2285 | 0.0390 | 1.2285 | 1.1084 | | 0.2613 | 18.8261 | 866 | 1.2261 | 0.0 | 1.2261 | 1.1073 | | 0.2613 | 18.8696 | 868 | 1.2744 | 0.0 | 1.2744 | 1.1289 | | 0.2613 | 18.9130 | 870 | 1.3796 | 0.2065 | 1.3796 | 1.1746 | | 0.2613 | 18.9565 | 872 | 1.4450 | 0.2065 | 1.4450 | 1.2021 | | 0.2613 | 19.0 | 874 | 1.4751 | 0.2126 | 1.4751 | 1.2146 | | 0.2613 | 19.0435 | 876 | 1.5427 | 0.2424 | 1.5427 | 1.2420 | | 0.2613 | 19.0870 | 878 | 1.5107 | 0.2424 | 1.5107 | 1.2291 | | 0.2613 | 19.1304 | 880 | 1.4398 | 0.2065 | 1.4398 | 1.1999 | | 0.2613 | 19.1739 | 882 | 1.3891 | 0.2065 | 1.3891 | 1.1786 | | 0.2613 | 19.2174 | 884 | 1.3882 | 0.2065 | 1.3882 | 1.1782 | | 0.2613 | 19.2609 | 886 | 1.3677 | 0.2372 | 1.3677 | 1.1695 | | 0.2613 | 19.3043 | 888 | 1.3458 | 0.2372 | 1.3458 | 1.1601 | | 0.2613 | 19.3478 | 890 | 1.3543 | 0.2065 | 1.3543 | 1.1638 | | 0.2613 | 19.3913 | 892 | 1.4147 | 0.2065 | 1.4147 | 1.1894 | | 0.2613 | 19.4348 | 894 | 1.4954 | 0.2065 | 1.4954 | 1.2229 | | 0.2613 | 19.4783 | 896 | 1.5624 | 0.2372 | 1.5624 | 1.2500 | | 0.2613 | 19.5217 | 898 | 1.5194 | 0.2065 | 1.5194 | 1.2326 | | 0.2613 | 19.5652 | 900 | 1.4135 | 0.1142 | 1.4135 | 1.1889 | | 0.2613 | 19.6087 | 902 | 1.3467 | 0.0760 | 1.3467 | 1.1605 | | 0.2613 | 19.6522 | 904 | 1.3329 | 0.0760 | 1.3329 | 1.1545 | | 0.2613 | 19.6957 | 906 | 1.3788 | 0.0760 | 1.3788 | 1.1742 | | 0.2613 | 19.7391 | 908 | 1.3941 | 0.1142 | 1.3941 | 1.1807 | | 0.2613 | 19.7826 | 910 | 1.3446 | 0.0760 | 1.3446 | 1.1596 | | 0.2613 | 19.8261 | 912 | 1.3464 | 0.0781 | 1.3464 | 1.1603 | | 0.2613 | 19.8696 | 914 | 1.3786 | 0.0781 | 1.3786 | 1.1741 | | 0.2613 | 19.9130 | 916 | 1.3890 | 0.0781 | 1.3890 | 1.1786 | | 0.2613 | 19.9565 | 918 | 1.4255 | 0.1142 | 1.4255 | 1.1940 | | 0.2613 | 20.0 | 920 | 1.4037 | 0.1142 | 1.4037 | 1.1848 | | 0.2613 | 20.0435 | 922 | 1.3277 | 0.0760 | 1.3277 | 1.1523 | | 0.2613 | 20.0870 | 924 | 1.3014 | 0.1700 | 1.3014 | 1.1408 | | 0.2613 | 20.1304 | 926 | 1.3892 | 0.2075 | 1.3892 | 1.1786 | | 0.2613 | 20.1739 | 928 | 1.5929 | 0.2317 | 1.5929 | 1.2621 | | 0.2613 | 20.2174 | 930 | 1.6970 | 0.2317 | 1.6970 | 1.3027 | | 0.2613 | 20.2609 | 932 | 1.7191 | 0.2363 | 1.7191 | 1.3112 | | 0.2613 | 20.3043 | 934 | 1.6680 | 0.2568 | 1.6680 | 1.2915 | | 0.2613 | 20.3478 | 936 | 1.6164 | 0.2239 | 1.6164 | 1.2714 | | 0.2613 | 20.3913 | 938 | 1.5876 | 0.2372 | 1.5876 | 1.2600 | | 0.2613 | 20.4348 | 940 | 1.5170 | 0.1744 | 1.5170 | 1.2317 | | 0.2613 | 20.4783 | 942 | 1.4762 | 0.0781 | 1.4762 | 1.2150 | | 0.2613 | 20.5217 | 944 | 1.4673 | 0.0781 | 1.4673 | 1.2113 | | 0.2613 | 20.5652 | 946 | 1.4335 | 0.0781 | 1.4335 | 1.1973 | | 0.2613 | 20.6087 | 948 | 1.4069 | 0.0390 | 1.4069 | 1.1861 | | 0.2613 | 20.6522 | 950 | 1.3963 | 0.0390 | 1.3963 | 1.1816 | | 0.2613 | 20.6957 | 952 | 1.3900 | 0.0781 | 1.3900 | 1.1790 | | 0.2613 | 20.7391 | 954 | 1.4040 | 0.0781 | 1.4040 | 1.1849 | | 0.2613 | 20.7826 | 956 | 1.4156 | 0.1142 | 1.4156 | 1.1898 | | 0.2613 | 20.8261 | 958 | 1.4111 | 0.2126 | 1.4111 | 1.1879 | | 0.2613 | 20.8696 | 960 | 1.4425 | 0.2474 | 1.4425 | 1.2011 | | 0.2613 | 20.9130 | 962 | 1.4642 | 0.2474 | 1.4642 | 1.2100 | | 0.2613 | 20.9565 | 964 | 1.4979 | 0.2474 | 1.4979 | 1.2239 | | 0.2613 | 21.0 | 966 | 1.4797 | 0.2474 | 1.4797 | 1.2164 | | 0.2613 | 21.0435 | 968 | 1.4519 | 0.0878 | 1.4519 | 1.2049 | | 0.2613 | 21.0870 | 970 | 1.4357 | 0.0401 | 1.4357 | 1.1982 | | 0.2613 | 21.1304 | 972 | 1.4458 | 0.0510 | 1.4458 | 1.2024 | | 0.2613 | 21.1739 | 974 | 1.4380 | 0.1228 | 1.4380 | 1.1992 | | 0.2613 | 21.2174 | 976 | 1.4340 | 0.0781 | 1.4340 | 1.1975 | | 0.2613 | 21.2609 | 978 | 1.4431 | 0.0781 | 1.4431 | 1.2013 | | 0.2613 | 21.3043 | 980 | 1.4430 | 0.0781 | 1.4430 | 1.2012 | | 0.2613 | 21.3478 | 982 | 1.4556 | 0.0781 | 1.4556 | 1.2065 | | 0.2613 | 21.3913 | 984 | 1.4817 | 0.1142 | 1.4817 | 1.2172 | | 0.2613 | 21.4348 | 986 | 1.4795 | 0.1142 | 1.4795 | 1.2164 | | 0.2613 | 21.4783 | 988 | 1.4291 | 0.0390 | 1.4291 | 1.1955 | | 0.2613 | 21.5217 | 990 | 1.3709 | 0.0 | 1.3709 | 1.1709 | | 0.2613 | 21.5652 | 992 | 1.3485 | 0.0 | 1.3485 | 1.1613 | | 0.2613 | 21.6087 | 994 | 1.3507 | 0.0 | 1.3507 | 1.1622 | | 0.2613 | 21.6522 | 996 | 1.3278 | 0.0 | 1.3278 | 1.1523 | | 0.2613 | 21.6957 | 998 | 1.3225 | 0.0 | 1.3225 | 1.1500 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1

thaffggg/94add4fb-1b6b-498e-bdda-09689212b56f

thaffggg

2025-01-20T23:40:46Z

6

0

peft

[ "peft", "safetensors", "phi3", "axolotl", "generated_from_trainer", "custom_code", "base_model:microsoft/Phi-3-mini-4k-instruct", "base_model:adapter:microsoft/Phi-3-mini-4k-instruct", "license:mit", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:10:03Z

--- library_name: peft license: mit base_model: microsoft/Phi-3-mini-4k-instruct tags: - axolotl - generated_from_trainer model-index: - name: 94add4fb-1b6b-498e-bdda-09689212b56f results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: microsoft/Phi-3-mini-4k-instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 2ae5e8a9f5305db8_train_data.json ds_type: json format: custom path: /workspace/input_data/2ae5e8a9f5305db8_train_data.json type: field_instruction: cleaned_description field_output: title format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: thaffggg/94add4fb-1b6b-498e-bdda-09689212b56f hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/2ae5e8a9f5305db8_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 5dae0642-01e7-4e34-8316-2fb97377e93c wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 5dae0642-01e7-4e34-8316-2fb97377e93c warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 94add4fb-1b6b-498e-bdda-09689212b56f This model is a fine-tuned version of [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.5853 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.4551 | 0.0023 | 200 | 0.5853 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

LHRuig/realfoto

LHRuig

2025-01-20T23:40:19Z

10

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:39:14Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # realfoto <Gallery /> ## Model description realfoto lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/realfoto/tree/main) them in the Files & versions tab.

thabel/whisper-medium-yo

thabel

2025-01-20T23:40:06Z

98

0

transformers

[ "transformers", "tensorboard", "safetensors", "whisper", "automatic-speech-recognition", "generated_from_trainer", "dv", "dataset:mozilla-foundation/common_voice_13_0", "base_model:openai/whisper-medium", "base_model:finetune:openai/whisper-medium", "license:apache-2.0", "model-index", "endpoints_compatible", "region:us" ]

automatic-speech-recognition

2024-12-20T11:34:37Z

--- library_name: transformers language: - dv license: apache-2.0 base_model: openai/whisper-medium tags: - generated_from_trainer datasets: - mozilla-foundation/common_voice_13_0 metrics: - wer model-index: - name: Whisper Small Dv - Sanchit Gandhi results: - task: name: Automatic Speech Recognition type: automatic-speech-recognition dataset: name: Common Voice 13 type: mozilla-foundation/common_voice_13_0 config: yo split: test args: yo metrics: - name: Wer type: wer value: 47.33077228772023 ---  # Whisper Small Dv - Sanchit Gandhi This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Common Voice 13 dataset. It achieves the following results on the evaluation set: - Loss: 0.8186 - Wer Ortho: 69.8365 - Wer: 47.3308 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: constant_with_warmup - lr_scheduler_warmup_steps: 50 - training_steps: 500 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer | |:-------------:|:------:|:----:|:---------------:|:---------:|:-------:| | 0.061 | 5.8824 | 500 | 0.8186 | 69.8365 | 47.3308 | ### Framework versions - Transformers 4.47.1 - Pytorch 2.5.1+cu121 - Datasets 3.2.0 - Tokenizers 0.21.0

mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF

mradermacher

2025-01-20T23:39:50Z

504

1

transformers

[ "transformers", "gguf", "text-generation-inference", "unsloth", "gemma2", "trl", "sft", "en", "id", "dataset:gmonsoon/CoT-id", "base_model:gmonsoon/Eunoia-Gemma-9B-o1-Indo", "base_model:quantized:gmonsoon/Eunoia-Gemma-9B-o1-Indo", "license:gemma", "endpoints_compatible", "region:us", "conversational" ]

null

2025-01-14T14:09:42Z

--- base_model: gmonsoon/Eunoia-Gemma-9B-o1-Indo datasets: - gmonsoon/CoT-id language: - en - id library_name: transformers license: gemma quantized_by: mradermacher tags: - text-generation-inference - transformers - unsloth - gemma2 - trl - sft --- ## About      static quants of https://huggingface.co/gmonsoon/Eunoia-Gemma-9B-o1-Indo  weighted/imatrix quants are available at https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q2_K.gguf) | Q2_K | 3.9 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q3_K_S.gguf) | Q3_K_S | 4.4 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q3_K_M.gguf) | Q3_K_M | 4.9 | lower quality | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q3_K_L.gguf) | Q3_K_L | 5.2 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.IQ4_XS.gguf) | IQ4_XS | 5.3 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q4_K_S.gguf) | Q4_K_S | 5.6 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q4_K_M.gguf) | Q4_K_M | 5.9 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q5_K_S.gguf) | Q5_K_S | 6.6 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q5_K_M.gguf) | Q5_K_M | 6.7 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q6_K.gguf) | Q6_K | 7.7 | very good quality | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.Q8_0.gguf) | Q8_0 | 9.9 | fast, best quality | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.f16.gguf) | f16 | 18.6 | 16 bpw, overkill | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time.

beichenxie/nikeai-test-2

beichenxie

2025-01-20T23:38:30Z

8

0

diffusers

[ "diffusers", "text-to-image", "diffusers-training", "lora", "replicate", "template:sd-lora", "sd3.5-large", "sd3.5", "sd3.5-diffusers", "base_model:stabilityai/stable-diffusion-3.5-large", "base_model:adapter:stabilityai/stable-diffusion-3.5-large", "license:other", "region:us" ]

text-to-image

2025-01-20T23:32:22Z

--- license: other library_name: diffusers tags: - text-to-image - diffusers-training - diffusers - lora - replicate - template:sd-lora - sd3.5-large - sd3.5 - sd3.5-diffusers base_model: stabilityai/stable-diffusion-3.5-large instance_prompt: HILOS, nikeai test widget: [] ---  # SD3.5-Large DreamBooth LoRA - beichenxie/nikeai-test-2 <Gallery /> ## Model description These are beichenxie/nikeai-test-2 DreamBooth LoRA weights for stable-diffusion-3.5-large. The weights were trained using [DreamBooth](https://dreambooth.github.io/) with the [SD3 diffusers trainer](https://github.com/huggingface/diffusers/blob/main/examples/dreambooth/README_sd3.md). Was LoRA for the text encoder enabled? False. ## Trigger words You should use `HILOS, nikeai test` to trigger the image generation. ## Download model [Download the *.safetensors LoRA](beichenxie/nikeai-test-2/tree/main) in the Files & versions tab. ## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers) ```py from diffusers import AutoPipelineForText2Image import torch pipeline = AutoPipelineForText2Image.from_pretrained(stable-diffusion-3.5-large, torch_dtype=torch.float16).to('cuda') pipeline.load_lora_weights('beichenxie/nikeai-test-2', weight_name='pytorch_lora_weights.safetensors') image = pipeline('HILOS, nikeai test').images[0] ``` ### Use it with UIs such as AUTOMATIC1111, Comfy UI, SD.Next, Invoke - **LoRA**: download **[`diffusers_lora_weights.safetensors` here 💾](/beichenxie/nikeai-test-2/blob/main/diffusers_lora_weights.safetensors)**. - Rename it and place it on your `models/Lora` folder. - On AUTOMATIC1111, load the LoRA by adding `<lora:your_new_name:1>` to your prompt. On ComfyUI just [load it as a regular LoRA](https://comfyanonymous.github.io/ComfyUI_examples/lora/). For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters) ## License Please adhere to the licensing terms as described [here](https://huggingface.co/stabilityai/stable-diffusion-3.5-large/blob/main/LICENSE.md). ## Training details Trained on Replicate using: [lucataco/stable-diffusion-3.5-large-lora-trainer](https://replicate.com/lucataco/stable-diffusion-3.5-large-lora-trainer) ## Intended uses & limitations #### How to use ```python # TODO: add an example code snippet for running this diffusion pipeline ``` #### Limitations and bias [TODO: provide examples of latent issues and potential remediations] ## Training details [TODO: describe the data used to train the model]

leixa/d711f2b7-a885-4745-987c-446462155057

leixa

2025-01-20T23:38:29Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:HuggingFaceM4/tiny-random-LlamaForCausalLM", "base_model:adapter:HuggingFaceM4/tiny-random-LlamaForCausalLM", "region:us" ]

null

2025-01-20T23:37:59Z

--- library_name: peft base_model: HuggingFaceM4/tiny-random-LlamaForCausalLM tags: - axolotl - generated_from_trainer model-index: - name: d711f2b7-a885-4745-987c-446462155057 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: HuggingFaceM4/tiny-random-LlamaForCausalLM bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 923c6e5b442f353f_train_data.json ds_type: json format: custom path: /workspace/input_data/923c6e5b442f353f_train_data.json type: field_input: confidence field_instruction: report field_output: statement format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: leixa/d711f2b7-a885-4745-987c-446462155057 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: 0 logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 100 micro_batch_size: 8 mlflow_experiment_name: /tmp/923c6e5b442f353f_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: techspear-hub wandb_mode: online wandb_name: 531f15f8-749c-479d-84e5-335387dc7e76 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 531f15f8-749c-479d-84e5-335387dc7e76 warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # d711f2b7-a885-4745-987c-446462155057 This model is a fine-tuned version of [HuggingFaceM4/tiny-random-LlamaForCausalLM](https://huggingface.co/HuggingFaceM4/tiny-random-LlamaForCausalLM) on the None dataset. It achieves the following results on the evaluation set: - Loss: 10.3697 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0042 | 1 | 10.3767 | | 10.3769 | 0.0379 | 9 | 10.3764 | | 10.3753 | 0.0758 | 18 | 10.3757 | | 10.3743 | 0.1137 | 27 | 10.3749 | | 10.3749 | 0.1516 | 36 | 10.3739 | | 10.3718 | 0.1895 | 45 | 10.3729 | | 10.3726 | 0.2274 | 54 | 10.3719 | | 10.3692 | 0.2653 | 63 | 10.3710 | | 10.3699 | 0.3032 | 72 | 10.3703 | | 10.369 | 0.3411 | 81 | 10.3699 | | 10.3684 | 0.3789 | 90 | 10.3698 | | 10.3699 | 0.4168 | 99 | 10.3697 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF

mradermacher

2025-01-20T23:38:02Z

573

1

transformers

[ "transformers", "gguf", "text-generation-inference", "unsloth", "gemma2", "trl", "sft", "en", "id", "dataset:gmonsoon/CoT-id", "base_model:gmonsoon/Eunoia-Gemma-9B-o1-Indo", "base_model:quantized:gmonsoon/Eunoia-Gemma-9B-o1-Indo", "license:gemma", "endpoints_compatible", "region:us", "imatrix", "conversational" ]

null

2025-01-17T09:56:31Z

--- base_model: gmonsoon/Eunoia-Gemma-9B-o1-Indo datasets: - gmonsoon/CoT-id language: - en - id library_name: transformers license: gemma quantized_by: mradermacher tags: - text-generation-inference - transformers - unsloth - gemma2 - trl - sft --- ## About      weighted/imatrix quants of https://huggingface.co/gmonsoon/Eunoia-Gemma-9B-o1-Indo  static quants are available at https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ1_S.gguf) | i1-IQ1_S | 2.5 | for the desperate | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ1_M.gguf) | i1-IQ1_M | 2.6 | mostly desperate | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 2.9 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ2_XS.gguf) | i1-IQ2_XS | 3.2 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ2_S.gguf) | i1-IQ2_S | 3.3 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ2_M.gguf) | i1-IQ2_M | 3.5 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q2_K_S.gguf) | i1-Q2_K_S | 3.7 | very low quality | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.9 | lower quality | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q2_K.gguf) | i1-Q2_K | 3.9 | IQ3_XXS probably better | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ3_XS.gguf) | i1-IQ3_XS | 4.2 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ3_S.gguf) | i1-IQ3_S | 4.4 | beats Q3_K* | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q3_K_S.gguf) | i1-Q3_K_S | 4.4 | IQ3_XS probably better | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ3_M.gguf) | i1-IQ3_M | 4.6 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.9 | IQ3_S probably better | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q3_K_L.gguf) | i1-Q3_K_L | 5.2 | IQ3_M probably better | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ4_XS.gguf) | i1-IQ4_XS | 5.3 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-IQ4_NL.gguf) | i1-IQ4_NL | 5.5 | prefer IQ4_XS | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q4_0.gguf) | i1-Q4_0 | 5.6 | fast, low quality | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q4_K_S.gguf) | i1-Q4_K_S | 5.6 | optimal size/speed/quality | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.9 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q4_1.gguf) | i1-Q4_1 | 6.1 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q5_K_S.gguf) | i1-Q5_K_S | 6.6 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q5_K_M.gguf) | i1-Q5_K_M | 6.7 | | | [GGUF](https://huggingface.co/mradermacher/Eunoia-Gemma-9B-o1-Indo-i1-GGUF/resolve/main/Eunoia-Gemma-9B-o1-Indo.i1-Q6_K.gguf) | i1-Q6_K | 7.7 | practically like static Q6_K | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.

nathanialhunt/a5b35d19-809b-4474-b3e7-2109f98d9b84

nathanialhunt

2025-01-20T23:38:00Z

8

0

peft

[ "peft", "safetensors", "mixtral", "axolotl", "generated_from_trainer", "base_model:Eurdem/Defne_llama3_2x8B", "base_model:adapter:Eurdem/Defne_llama3_2x8B", "license:llama3", "region:us" ]

null

2025-01-20T22:39:16Z

--- library_name: peft license: llama3 base_model: Eurdem/Defne_llama3_2x8B tags: - axolotl - generated_from_trainer model-index: - name: a5b35d19-809b-4474-b3e7-2109f98d9b84 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Eurdem/Defne_llama3_2x8B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 7e3b096caae7eb1c_train_data.json ds_type: json format: custom path: /workspace/input_data/7e3b096caae7eb1c_train_data.json type: field_instruction: instruction field_output: output format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: nathanialhunt/a5b35d19-809b-4474-b3e7-2109f98d9b84 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/7e3b096caae7eb1c_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 319fd169-6b6a-48ba-95dd-c64f160d3657 wandb_project: Birthday-SN56-24-Gradients-On-Demand wandb_run: your_name wandb_runid: 319fd169-6b6a-48ba-95dd-c64f160d3657 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # a5b35d19-809b-4474-b3e7-2109f98d9b84 This model is a fine-tuned version of [Eurdem/Defne_llama3_2x8B](https://huggingface.co/Eurdem/Defne_llama3_2x8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.0000 | 1 | nan | | 0.0 | 0.0001 | 3 | nan | | 0.0 | 0.0002 | 6 | nan | | 0.0 | 0.0002 | 9 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

lesso10/a4fd6254-1a96-4603-aea9-5a2d20a2f3bc

lesso10

2025-01-20T23:37:34Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2.5-Coder-7B-Instruct", "base_model:adapter:unsloth/Qwen2.5-Coder-7B-Instruct", "license:apache-2.0", "4-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:55:41Z

--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2.5-Coder-7B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: a4fd6254-1a96-4603-aea9-5a2d20a2f3bc results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2.5-Coder-7B-Instruct bf16: auto chat_template: llama3 datasets: - data_files: - 5da4fdb4f9d40cf6_train_data.json ds_type: json format: custom path: /workspace/input_data/5da4fdb4f9d40cf6_train_data.json type: field_input: topic; field_instruction: message_1 field_output: message_2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: true gradient_checkpointing: true group_by_length: false hub_model_id: lesso10/a4fd6254-1a96-4603-aea9-5a2d20a2f3bc hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: true load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/5da4fdb4f9d40cf6_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 971c59eb-5de8-4a78-8d22-6a7da4c9ee82 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 971c59eb-5de8-4a78-8d22-6a7da4c9ee82 warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # a4fd6254-1a96-4603-aea9-5a2d20a2f3bc This model is a fine-tuned version of [unsloth/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0001 | 1 | nan | | 0.0 | 0.0005 | 5 | nan | | 0.0 | 0.0011 | 10 | nan | | 0.0 | 0.0016 | 15 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

LHRuig/detailsharp

LHRuig

2025-01-20T23:37:03Z

28

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:36:29Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # detailsharp <Gallery /> ## Model description detailsharp lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/detailsharp/tree/main) them in the Files & versions tab.

matrixportal/Llama-3.1-8B-BookAdventures-GGUF

matrixportal

2025-01-20T23:36:24Z

28

0

null

[ "gguf", "llama-factory", "full", "generated_from_trainer", "llama-cpp", "gguf-my-repo", "base_model:KoboldAI/Llama-3.1-8B-BookAdventures", "base_model:quantized:KoboldAI/Llama-3.1-8B-BookAdventures", "license:cc-by-nc-sa-4.0", "endpoints_compatible", "region:us", "conversational" ]

null

2025-01-20T23:35:43Z

--- license: cc-by-nc-sa-4.0 base_model: KoboldAI/Llama-3.1-8B-BookAdventures tags: - llama-factory - full - generated_from_trainer - llama-cpp - gguf-my-repo model-index: - name: KoboldAI/Llama-3.1-8B-BookAdventures results: [] --- # matrixportal/Llama-3.1-8B-BookAdventures-GGUF This model was converted to GGUF format from [`KoboldAI/Llama-3.1-8B-BookAdventures`](https://huggingface.co/KoboldAI/Llama-3.1-8B-BookAdventures) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space. Refer to the [original model card](https://huggingface.co/KoboldAI/Llama-3.1-8B-BookAdventures) for more details on the model. ## Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) ```bash brew install llama.cpp ``` Invoke the llama.cpp server or the CLI. ### CLI: ```bash llama-cli --hf-repo matrixportal/Llama-3.1-8B-BookAdventures-GGUF --hf-file llama-3.1-8b-bookadventures-q4_k_m.gguf -p "The meaning to life and the universe is" ``` ### Server: ```bash llama-server --hf-repo matrixportal/Llama-3.1-8B-BookAdventures-GGUF --hf-file llama-3.1-8b-bookadventures-q4_k_m.gguf -c 2048 ``` Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well. Step 1: Clone llama.cpp from GitHub. ``` git clone https://github.com/ggerganov/llama.cpp ``` Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux). ``` cd llama.cpp && LLAMA_CURL=1 make ``` Step 3: Run inference through the main binary. ``` ./llama-cli --hf-repo matrixportal/Llama-3.1-8B-BookAdventures-GGUF --hf-file llama-3.1-8b-bookadventures-q4_k_m.gguf -p "The meaning to life and the universe is" ``` or ``` ./llama-server --hf-repo matrixportal/Llama-3.1-8B-BookAdventures-GGUF --hf-file llama-3.1-8b-bookadventures-q4_k_m.gguf -c 2048 ```

fpadovani/english_childes_context_42

fpadovani

2025-01-20T23:36:06Z

6

0

transformers

[ "transformers", "safetensors", "roberta", "fill-mask", "generated_from_trainer", "autotrain_compatible", "endpoints_compatible", "region:us" ]

fill-mask

2025-01-20T22:20:37Z

--- library_name: transformers tags: - generated_from_trainer model-index: - name: childes_mlm_unmasking_context results: [] ---  # childes_mlm_unmasking_context This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 2.1318 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 100000 - training_steps: 400000 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-------:|:-----:|:---------------:| | No log | 1.2698 | 2000 | 5.5274 | | 6.237 | 2.5397 | 4000 | 5.4702 | | 6.237 | 3.8095 | 6000 | 5.3851 | | 5.4081 | 5.0794 | 8000 | 5.3058 | | 5.4081 | 6.3492 | 10000 | 4.0259 | | 4.2991 | 7.6190 | 12000 | 3.3054 | | 4.2991 | 8.8889 | 14000 | 3.0169 | | 3.104 | 10.1587 | 16000 | 2.7758 | | 3.104 | 11.4286 | 18000 | 2.6658 | | 2.7387 | 12.6984 | 20000 | 2.5578 | | 2.7387 | 13.9683 | 22000 | 2.5101 | | 2.5352 | 15.2381 | 24000 | 2.4254 | | 2.5352 | 16.5079 | 26000 | 2.3459 | | 2.4005 | 17.7778 | 28000 | 2.3221 | | 2.4005 | 19.0476 | 30000 | 2.2965 | | 2.3081 | 20.3175 | 32000 | 2.2853 | | 2.3081 | 21.5873 | 34000 | 2.2284 | | 2.2426 | 22.8571 | 36000 | 2.1734 | | 2.2426 | 24.1270 | 38000 | 2.1705 | | 2.193 | 25.3968 | 40000 | 2.1931 | | 2.193 | 26.6667 | 42000 | 2.1680 | | 2.1678 | 27.9365 | 44000 | 2.1337 | | 2.1678 | 29.2063 | 46000 | 2.1411 | | 2.1464 | 30.4762 | 48000 | 2.1310 | | 2.1464 | 31.7460 | 50000 | 2.1229 | | 2.1285 | 33.0159 | 52000 | 2.1331 | | 2.1285 | 34.2857 | 54000 | 2.1592 | | 2.1199 | 35.5556 | 56000 | 2.1318 | ### Framework versions - Transformers 4.45.2 - Pytorch 2.5.1+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

dimasik1987/d0b444d3-4c6e-4c6f-a3e0-c745e20ca3f8

dimasik1987

2025-01-20T23:36:04Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:trl-internal-testing/tiny-random-LlamaForCausalLM", "base_model:adapter:trl-internal-testing/tiny-random-LlamaForCausalLM", "region:us" ]

null

2025-01-20T23:35:41Z

--- library_name: peft base_model: trl-internal-testing/tiny-random-LlamaForCausalLM tags: - axolotl - generated_from_trainer model-index: - name: d0b444d3-4c6e-4c6f-a3e0-c745e20ca3f8 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: trl-internal-testing/tiny-random-LlamaForCausalLM bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - dd22c8863ed4176b_train_data.json ds_type: json format: custom path: /workspace/input_data/dd22c8863ed4176b_train_data.json type: field_input: text field_instruction: title field_output: summary format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: dimasik1987/d0b444d3-4c6e-4c6f-a3e0-c745e20ca3f8 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/dd22c8863ed4176b_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ee0747e5-378f-43ac-83d3-8dd08d6876bf wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ee0747e5-378f-43ac-83d3-8dd08d6876bf warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # d0b444d3-4c6e-4c6f-a3e0-c745e20ca3f8 This model is a fine-tuned version of [trl-internal-testing/tiny-random-LlamaForCausalLM](https://huggingface.co/trl-internal-testing/tiny-random-LlamaForCausalLM) on the None dataset. It achieves the following results on the evaluation set: - Loss: 10.3303 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0292 | 1 | 10.3601 | | 10.3594 | 0.1460 | 5 | 10.3562 | | 10.3515 | 0.2920 | 10 | 10.3464 | | 10.3409 | 0.4380 | 15 | 10.3383 | | 10.337 | 0.5839 | 20 | 10.3330 | | 10.3318 | 0.7299 | 25 | 10.3307 | | 10.3292 | 0.8759 | 30 | 10.3303 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

LHRuig/moodcine

LHRuig

2025-01-20T23:35:59Z

12

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:35:47Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # moodcine <Gallery /> ## Model description moodcine lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/moodcine/tree/main) them in the Files & versions tab.

prxy5604/37578cc4-aa06-437b-80e1-561bf03536ef

prxy5604

2025-01-20T23:35:57Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:trl-internal-testing/tiny-random-LlamaForCausalLM", "base_model:adapter:trl-internal-testing/tiny-random-LlamaForCausalLM", "region:us" ]

null

2025-01-20T23:35:36Z

--- library_name: peft base_model: trl-internal-testing/tiny-random-LlamaForCausalLM tags: - axolotl - generated_from_trainer model-index: - name: 37578cc4-aa06-437b-80e1-561bf03536ef results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: trl-internal-testing/tiny-random-LlamaForCausalLM bf16: true chat_template: llama3 data_processes: 16 dataset_prepared_path: null datasets: - data_files: - dd22c8863ed4176b_train_data.json ds_type: json format: custom path: /workspace/input_data/dd22c8863ed4176b_train_data.json type: field_input: text field_instruction: title field_output: summary format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto do_eval: true early_stopping_patience: 5 eval_batch_size: 4 eval_max_new_tokens: 128 eval_steps: 50 eval_table_size: null evals_per_epoch: null flash_attention: true fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: prxy5604/37578cc4-aa06-437b-80e1-561bf03536ef hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 128 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 64 lora_target_linear: true lr_scheduler: cosine max_grad_norm: 1.0 max_memory: 0: 75GB max_steps: 200 micro_batch_size: 8 mlflow_experiment_name: /tmp/dd22c8863ed4176b_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 50 saves_per_epoch: null sequence_len: 1024 strict: false tf32: true tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ee0747e5-378f-43ac-83d3-8dd08d6876bf wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ee0747e5-378f-43ac-83d3-8dd08d6876bf warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 37578cc4-aa06-437b-80e1-561bf03536ef This model is a fine-tuned version of [trl-internal-testing/tiny-random-LlamaForCausalLM](https://huggingface.co/trl-internal-testing/tiny-random-LlamaForCausalLM) on the None dataset. It achieves the following results on the evaluation set: - Loss: 10.2951 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 8 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 52 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 10.3598 | 0.0580 | 1 | 10.3596 | | 10.5029 | 2.8986 | 50 | 10.2951 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

JacksonBrune/919b9850-bcff-4436-97bc-01c41a6c1517

JacksonBrune

2025-01-20T23:35:13Z

6

0

peft

[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "base_model:Intel/neural-chat-7b-v3-3", "base_model:adapter:Intel/neural-chat-7b-v3-3", "license:apache-2.0", "region:us" ]

null

2025-01-20T20:14:32Z

--- library_name: peft license: apache-2.0 base_model: Intel/neural-chat-7b-v3-3 tags: - axolotl - generated_from_trainer model-index: - name: 919b9850-bcff-4436-97bc-01c41a6c1517 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Intel/neural-chat-7b-v3-3 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 50587f38ed52161f_train_data.json ds_type: json format: custom path: /workspace/input_data/50587f38ed52161f_train_data.json type: field_input: file field_instruction: directory field_output: content format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: JacksonBrune/919b9850-bcff-4436-97bc-01c41a6c1517 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 10 micro_batch_size: 2 mlflow_experiment_name: /tmp/50587f38ed52161f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 3f448567-d842-4770-a570-924da07f2a6c wandb_project: Birthday-SN56-12-Gradients-On-Demand wandb_run: your_name wandb_runid: 3f448567-d842-4770-a570-924da07f2a6c warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 919b9850-bcff-4436-97bc-01c41a6c1517 This model is a fine-tuned version of [Intel/neural-chat-7b-v3-3](https://huggingface.co/Intel/neural-chat-7b-v3-3) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.0000 | 1 | nan | | 0.0 | 0.0000 | 3 | nan | | 0.0 | 0.0000 | 6 | nan | | 0.0 | 0.0000 | 9 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

mradermacher/NBeerbower-ConversationalMix-8b-GGUF

mradermacher

2025-01-20T23:34:47Z

328

1

transformers

[ "transformers", "gguf", "mergekit", "merge", "en", "base_model:MrRobotoAI/NBeerbower-ConversationalMix-8b", "base_model:quantized:MrRobotoAI/NBeerbower-ConversationalMix-8b", "endpoints_compatible", "region:us", "conversational" ]

null

2025-01-19T21:42:40Z

--- base_model: MrRobotoAI/NBeerbower-ConversationalMix-8b language: - en library_name: transformers quantized_by: mradermacher tags: - mergekit - merge --- ## About      static quants of https://huggingface.co/MrRobotoAI/NBeerbower-ConversationalMix-8b  weighted/imatrix quants are available at https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-i1-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q2_K.gguf) | Q2_K | 3.3 | | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q3_K_S.gguf) | Q3_K_S | 3.8 | | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q3_K_M.gguf) | Q3_K_M | 4.1 | lower quality | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q3_K_L.gguf) | Q3_K_L | 4.4 | | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.IQ4_XS.gguf) | IQ4_XS | 4.6 | | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q4_K_S.gguf) | Q4_K_S | 4.8 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q4_K_M.gguf) | Q4_K_M | 5.0 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q5_K_S.gguf) | Q5_K_S | 5.7 | | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q5_K_M.gguf) | Q5_K_M | 5.8 | | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q6_K.gguf) | Q6_K | 6.7 | very good quality | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.Q8_0.gguf) | Q8_0 | 8.6 | fast, best quality | | [GGUF](https://huggingface.co/mradermacher/NBeerbower-ConversationalMix-8b-GGUF/resolve/main/NBeerbower-ConversationalMix-8b.f16.gguf) | f16 | 16.2 | 16 bpw, overkill | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time.

lesso13/d9c12229-8dbb-45fa-ba4f-9717c95db112

lesso13

2025-01-20T23:34:35Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "base_model:adapter:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:03:26Z

--- library_name: peft base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f tags: - axolotl - generated_from_trainer model-index: - name: d9c12229-8dbb-45fa-ba4f-9717c95db112 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f bf16: true chat_template: llama3 datasets: - data_files: - train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json ds_type: json format: custom path: /workspace/input_data/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso13/d9c12229-8dbb-45fa-ba4f-9717c95db112 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 45GiB max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075526eb-32e0-4485-aab7-014e4d302171 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075526eb-32e0-4485-aab7-014e4d302171 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # d9c12229-8dbb-45fa-ba4f-9717c95db112 This model is a fine-tuned version of [samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f](https://huggingface.co/samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.0002 | 1 | nan | | 0.0 | 0.0010 | 5 | nan | | 0.0 | 0.0020 | 10 | nan | | 0.0 | 0.0031 | 15 | nan | | 0.0 | 0.0041 | 20 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

demohong/7f82ae83-b307-42ac-b3a7-914fcf126608

demohong

2025-01-20T23:30:24Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:unsloth/Meta-Llama-3.1-8B", "base_model:adapter:unsloth/Meta-Llama-3.1-8B", "license:llama3.1", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:59:26Z

--- library_name: peft license: llama3.1 base_model: unsloth/Meta-Llama-3.1-8B tags: - axolotl - generated_from_trainer model-index: - name: 7f82ae83-b307-42ac-b3a7-914fcf126608 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Meta-Llama-3.1-8B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 91e193d3dca1611f_train_data.json ds_type: json format: custom path: /workspace/input_data/91e193d3dca1611f_train_data.json type: field_input: parent_id field_instruction: role field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: demohong/7f82ae83-b307-42ac-b3a7-914fcf126608 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/91e193d3dca1611f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 856a9aac-189f-40f7-b27c-c5616995b0d1 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 856a9aac-189f-40f7-b27c-c5616995b0d1 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 7f82ae83-b307-42ac-b3a7-914fcf126608 This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.5966 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.3105 | 0.0705 | 200 | 1.5966 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

nbninh/d2c84520-d1af-4aa3-b9a1-d0cdf04d4d4d

nbninh

2025-01-20T23:29:25Z

8

0

peft

[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "base_model:unsloth/Phi-3-medium-4k-instruct", "base_model:adapter:unsloth/Phi-3-medium-4k-instruct", "license:mit", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T21:45:32Z

--- library_name: peft license: mit base_model: unsloth/Phi-3-medium-4k-instruct tags: - axolotl - generated_from_trainer model-index: - name: d2c84520-d1af-4aa3-b9a1-d0cdf04d4d4d results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Phi-3-medium-4k-instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 41d403c8b37c92fc_train_data.json ds_type: json format: custom path: /workspace/input_data/41d403c8b37c92fc_train_data.json type: field_input: mesh_terms field_instruction: title field_output: abstract format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nbninh/d2c84520-d1af-4aa3-b9a1-d0cdf04d4d4d hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/41d403c8b37c92fc_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075ea541-bd04-429e-a989-c49dabc36fc3 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075ea541-bd04-429e-a989-c49dabc36fc3 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # d2c84520-d1af-4aa3-b9a1-d0cdf04d4d4d This model is a fine-tuned version of [unsloth/Phi-3-medium-4k-instruct](https://huggingface.co/unsloth/Phi-3-medium-4k-instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.4790 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 5.7635 | 0.0048 | 200 | 1.4790 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

LHRuig/detail2k

LHRuig

2025-01-20T23:29:11Z

5

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:28:58Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # detail2k <Gallery /> ## Model description detail2k lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/detail2k/tree/main) them in the Files & versions tab.

MikeRoz/deepseek-ai_DeepSeek-R1-Distill-Llama-70B-4.25bpw-h6-exl2

MikeRoz

2025-01-20T23:28:11Z

227

4

null

[ "safetensors", "llama", "exl2", "region:us" ]

null

2025-01-20T20:50:11Z

# DeepSeek-R1    <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- **Post-Training: Large-Scale Reinforcement Learning on the Base Model** - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- **Distillation: Smaller Models Can Be Powerful Too** - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> | **Model** | **#Total Params** | **#Activated Params** | **Context Length** | **Download** | | :------------: | :------------: | :------------: | :------------: | :------------: | | DeepSeek-R1-Zero | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) | | DeepSeek-R1 | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) | </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> | **Model** | **Base Model** | **Download** | | :------------: | :------------: | :------------: | | DeepSeek-R1-Distill-Qwen-1.5B | [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) | | DeepSeek-R1-Distill-Qwen-7B | [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) | | DeepSeek-R1-Distill-Llama-8B | [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) | | DeepSeek-R1-Distill-Qwen-14B | [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) | |DeepSeek-R1-Distill-Qwen-32B | [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) | | DeepSeek-R1-Distill-Llama-70B | [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) | </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> | Category | Benchmark (Metric) | Claude-3.5-Sonnet-1022 | GPT-4o 0513 | DeepSeek V3 | OpenAI o1-mini | OpenAI o1-1217 | DeepSeek R1 | |----------|-------------------|----------------------|------------|--------------|----------------|------------|--------------| | | Architecture | - | - | MoE | - | - | MoE | | | # Activated Params | - | - | 37B | - | - | 37B | | | # Total Params | - | - | 671B | - | - | 671B | | English | MMLU (Pass@1) | 88.3 | 87.2 | 88.5 | 85.2 | **91.8** | 90.8 | | | MMLU-Redux (EM) | 88.9 | 88.0 | 89.1 | 86.7 | - | **92.9** | | | MMLU-Pro (EM) | 78.0 | 72.6 | 75.9 | 80.3 | - | **84.0** | | | DROP (3-shot F1) | 88.3 | 83.7 | 91.6 | 83.9 | 90.2 | **92.2** | | | IF-Eval (Prompt Strict) | **86.5** | 84.3 | 86.1 | 84.8 | - | 83.3 | | | GPQA-Diamond (Pass@1) | 65.0 | 49.9 | 59.1 | 60.0 | **75.7** | 71.5 | | | SimpleQA (Correct) | 28.4 | 38.2 | 24.9 | 7.0 | **47.0** | 30.1 | | | FRAMES (Acc.) | 72.5 | 80.5 | 73.3 | 76.9 | - | **82.5** | | | AlpacaEval2.0 (LC-winrate) | 52.0 | 51.1 | 70.0 | 57.8 | - | **87.6** | | | ArenaHard (GPT-4-1106) | 85.2 | 80.4 | 85.5 | 92.0 | - | **92.3** | | Code | LiveCodeBench (Pass@1-COT) | 33.8 | 34.2 | - | 53.8 | 63.4 | **65.9** | | | Codeforces (Percentile) | 20.3 | 23.6 | 58.7 | 93.4 | **96.6** | 96.3 | | | Codeforces (Rating) | 717 | 759 | 1134 | 1820 | **2061** | 2029 | | | SWE Verified (Resolved) | **50.8** | 38.8 | 42.0 | 41.6 | 48.9 | 49.2 | | | Aider-Polyglot (Acc.) | 45.3 | 16.0 | 49.6 | 32.9 | **61.7** | 53.3 | | Math | AIME 2024 (Pass@1) | 16.0 | 9.3 | 39.2 | 63.6 | 79.2 | **79.8** | | | MATH-500 (Pass@1) | 78.3 | 74.6 | 90.2 | 90.0 | 96.4 | **97.3** | | | CNMO 2024 (Pass@1) | 13.1 | 10.8 | 43.2 | 67.6 | - | **78.8** | | Chinese | CLUEWSC (EM) | 85.4 | 87.9 | 90.9 | 89.9 | - | **92.8** | | | C-Eval (EM) | 76.7 | 76.0 | 86.5 | 68.9 | - | **91.8** | | | C-SimpleQA (Correct) | 55.4 | 58.7 | **68.0** | 40.3 | - | 63.7 | </div> ### Distilled Model Evaluation <div align="center"> | Model | AIME 2024 pass@1 | AIME 2024 cons@64 | MATH-500 pass@1 | GPQA Diamond pass@1 | LiveCodeBench pass@1 | CodeForces rating | |------------------------------------------|------------------|-------------------|-----------------|----------------------|----------------------|-------------------| | GPT-4o-0513 | 9.3 | 13.4 | 74.6 | 49.9 | 32.9 | 759 | | Claude-3.5-Sonnet-1022 | 16.0 | 26.7 | 78.3 | 65.0 | 38.9 | 717 | | o1-mini | 63.6 | 80.0 | 90.0 | 60.0 | 53.8 | **1820** | | QwQ-32B-Preview | 44.0 | 60.0 | 90.6 | 54.5 | 41.9 | 1316 | | DeepSeek-R1-Distill-Qwen-1.5B | 28.9 | 52.7 | 83.9 | 33.8 | 16.9 | 954 | | DeepSeek-R1-Distill-Qwen-7B | 55.5 | 83.3 | 92.8 | 49.1 | 37.6 | 1189 | | DeepSeek-R1-Distill-Qwen-14B | 69.7 | 80.0 | 93.9 | 59.1 | 53.1 | 1481 | | DeepSeek-R1-Distill-Qwen-32B | **72.6** | 83.3 | 94.3 | 62.1 | 57.2 | 1691 | | DeepSeek-R1-Distill-Llama-8B | 50.4 | 80.0 | 89.1 | 49.0 | 39.6 | 1205 | | DeepSeek-R1-Distill-Llama-70B | 70.0 | **86.7** | **94.5** | **65.2** | **57.5** | 1633 | </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` **NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output.** ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).

LHRuig/adddetails

LHRuig

2025-01-20T23:26:11Z

9

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:26:06Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # adddetail <Gallery /> ## Model description adddetail lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/adddetails/tree/main) them in the Files & versions tab.

lesso10/1c6a64b4-e7f1-4e45-9efb-b4b0d937a28a

lesso10

2025-01-20T23:25:07Z

8

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/CodeLlama-13b-hf-flash", "base_model:adapter:NousResearch/CodeLlama-13b-hf-flash", "4-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:41:43Z

--- library_name: peft base_model: NousResearch/CodeLlama-13b-hf-flash tags: - axolotl - generated_from_trainer model-index: - name: 1c6a64b4-e7f1-4e45-9efb-b4b0d937a28a results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/CodeLlama-13b-hf-flash bf16: auto chat_template: llama3 datasets: - data_files: - 9ff4e3b24bf3b2a4_train_data.json ds_type: json format: custom path: /workspace/input_data/9ff4e3b24bf3b2a4_train_data.json type: field_input: sentence1 field_instruction: phrase1 field_output: sentence2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: true gradient_checkpointing: true group_by_length: false hub_model_id: lesso10/1c6a64b4-e7f1-4e45-9efb-b4b0d937a28a hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: true load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/9ff4e3b24bf3b2a4_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # 1c6a64b4-e7f1-4e45-9efb-b4b0d937a28a This model is a fine-tuned version of [NousResearch/CodeLlama-13b-hf-flash](https://huggingface.co/NousResearch/CodeLlama-13b-hf-flash) on the None dataset. It achieves the following results on the evaluation set: - Loss: 3.0292 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0003 | 1 | 4.9637 | | 5.0443 | 0.0014 | 5 | 4.8203 | | 4.2709 | 0.0029 | 10 | 3.9057 | | 3.0327 | 0.0043 | 15 | 3.1180 | | 3.3112 | 0.0057 | 20 | 3.0652 | | 2.8835 | 0.0071 | 25 | 3.0292 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

John6666/rippillustrious-v10-sdxl

John6666

2025-01-20T23:24:26Z

81

0

diffusers

[ "diffusers", "safetensors", "text-to-image", "stable-diffusion", "stable-diffusion-xl", "anime", "girls", "illustrious", "en", "base_model:OnomaAIResearch/Illustrious-xl-early-release-v0", "base_model:finetune:OnomaAIResearch/Illustrious-xl-early-release-v0", "license:other", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionXLPipeline", "region:us" ]

text-to-image

2025-01-20T23:18:28Z

--- license: other license_name: faipl-1.0-sd license_link: https://freedevproject.org/faipl-1.0-sd/ language: - en library_name: diffusers pipeline_tag: text-to-image tags: - text-to-image - stable-diffusion - stable-diffusion-xl - anime - girls - illustrious base_model: OnomaAIResearch/Illustrious-xl-early-release-v0 --- Original model is [here](https://civitai.com/models/1163094/rippillustrious?modelVersionId=1308356). This model created by [tbets182132](https://civitai.com/user/tbets182132).

LHRuig/kodakmotion

LHRuig

2025-01-20T23:24:01Z

11

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:23:17Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # kodakmotion <Gallery /> ## Model description kodakmotion lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/kodakmotion/tree/main) them in the Files & versions tab.

lhong4759/bbef3a84-7d4f-465d-914c-e5286aa7e060

lhong4759

2025-01-20T23:23:56Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "base_model:adapter:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:03:00Z

--- library_name: peft base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f tags: - axolotl - generated_from_trainer model-index: - name: bbef3a84-7d4f-465d-914c-e5286aa7e060 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json ds_type: json format: custom path: /workspace/input_data/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: lhong4759/bbef3a84-7d4f-465d-914c-e5286aa7e060 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075526eb-32e0-4485-aab7-014e4d302171 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075526eb-32e0-4485-aab7-014e4d302171 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # bbef3a84-7d4f-465d-914c-e5286aa7e060 This model is a fine-tuned version of [samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f](https://huggingface.co/samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.1312 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.2357 | 0.0407 | 200 | 1.1312 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

adamo1139/DeepSeek-R1-Distill-Qwen-1.5B-3bpw-exl2

adamo1139

2025-01-20T23:23:37Z

5

0

null

[ "qwen2", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "3-bit", "exl2", "region:us" ]

null

2025-01-20T22:53:24Z

--- base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --- # DeepSeek-R1    <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- **Post-Training: Large-Scale Reinforcement Learning on the Base Model** - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- **Distillation: Smaller Models Can Be Powerful Too** - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> | **Model** | **#Total Params** | **#Activated Params** | **Context Length** | **Download** | | :------------: | :------------: | :------------: | :------------: | :------------: | | DeepSeek-R1-Zero | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) | | DeepSeek-R1 | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) | </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> | **Model** | **Base Model** | **Download** | | :------------: | :------------: | :------------: | | DeepSeek-R1-Distill-Qwen-1.5B | [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) | | DeepSeek-R1-Distill-Qwen-7B | [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) | | DeepSeek-R1-Distill-Llama-8B | [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) | | DeepSeek-R1-Distill-Qwen-14B | [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) | |DeepSeek-R1-Distill-Qwen-32B | [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) | | DeepSeek-R1-Distill-Llama-70B | [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) | </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> | Category | Benchmark (Metric) | Claude-3.5-Sonnet-1022 | GPT-4o 0513 | DeepSeek V3 | OpenAI o1-mini | OpenAI o1-1217 | DeepSeek R1 | |----------|-------------------|----------------------|------------|--------------|----------------|------------|--------------| | | Architecture | - | - | MoE | - | - | MoE | | | # Activated Params | - | - | 37B | - | - | 37B | | | # Total Params | - | - | 671B | - | - | 671B | | English | MMLU (Pass@1) | 88.3 | 87.2 | 88.5 | 85.2 | **91.8** | 90.8 | | | MMLU-Redux (EM) | 88.9 | 88.0 | 89.1 | 86.7 | - | **92.9** | | | MMLU-Pro (EM) | 78.0 | 72.6 | 75.9 | 80.3 | - | **84.0** | | | DROP (3-shot F1) | 88.3 | 83.7 | 91.6 | 83.9 | 90.2 | **92.2** | | | IF-Eval (Prompt Strict) | **86.5** | 84.3 | 86.1 | 84.8 | - | 83.3 | | | GPQA-Diamond (Pass@1) | 65.0 | 49.9 | 59.1 | 60.0 | **75.7** | 71.5 | | | SimpleQA (Correct) | 28.4 | 38.2 | 24.9 | 7.0 | **47.0** | 30.1 | | | FRAMES (Acc.) | 72.5 | 80.5 | 73.3 | 76.9 | - | **82.5** | | | AlpacaEval2.0 (LC-winrate) | 52.0 | 51.1 | 70.0 | 57.8 | - | **87.6** | | | ArenaHard (GPT-4-1106) | 85.2 | 80.4 | 85.5 | 92.0 | - | **92.3** | | Code | LiveCodeBench (Pass@1-COT) | 33.8 | 34.2 | - | 53.8 | 63.4 | **65.9** | | | Codeforces (Percentile) | 20.3 | 23.6 | 58.7 | 93.4 | **96.6** | 96.3 | | | Codeforces (Rating) | 717 | 759 | 1134 | 1820 | **2061** | 2029 | | | SWE Verified (Resolved) | **50.8** | 38.8 | 42.0 | 41.6 | 48.9 | 49.2 | | | Aider-Polyglot (Acc.) | 45.3 | 16.0 | 49.6 | 32.9 | **61.7** | 53.3 | | Math | AIME 2024 (Pass@1) | 16.0 | 9.3 | 39.2 | 63.6 | 79.2 | **79.8** | | | MATH-500 (Pass@1) | 78.3 | 74.6 | 90.2 | 90.0 | 96.4 | **97.3** | | | CNMO 2024 (Pass@1) | 13.1 | 10.8 | 43.2 | 67.6 | - | **78.8** | | Chinese | CLUEWSC (EM) | 85.4 | 87.9 | 90.9 | 89.9 | - | **92.8** | | | C-Eval (EM) | 76.7 | 76.0 | 86.5 | 68.9 | - | **91.8** | | | C-SimpleQA (Correct) | 55.4 | 58.7 | **68.0** | 40.3 | - | 63.7 | </div> ### Distilled Model Evaluation <div align="center"> | Model | AIME 2024 pass@1 | AIME 2024 cons@64 | MATH-500 pass@1 | GPQA Diamond pass@1 | LiveCodeBench pass@1 | CodeForces rating | |------------------------------------------|------------------|-------------------|-----------------|----------------------|----------------------|-------------------| | GPT-4o-0513 | 9.3 | 13.4 | 74.6 | 49.9 | 32.9 | 759 | | Claude-3.5-Sonnet-1022 | 16.0 | 26.7 | 78.3 | 65.0 | 38.9 | 717 | | o1-mini | 63.6 | 80.0 | 90.0 | 60.0 | 53.8 | **1820** | | QwQ-32B-Preview | 44.0 | 60.0 | 90.6 | 54.5 | 41.9 | 1316 | | DeepSeek-R1-Distill-Qwen-1.5B | 28.9 | 52.7 | 83.9 | 33.8 | 16.9 | 954 | | DeepSeek-R1-Distill-Qwen-7B | 55.5 | 83.3 | 92.8 | 49.1 | 37.6 | 1189 | | DeepSeek-R1-Distill-Qwen-14B | 69.7 | 80.0 | 93.9 | 59.1 | 53.1 | 1481 | | DeepSeek-R1-Distill-Qwen-32B | **72.6** | 83.3 | 94.3 | 62.1 | 57.2 | 1691 | | DeepSeek-R1-Distill-Llama-8B | 50.4 | 80.0 | 89.1 | 49.0 | 39.6 | 1205 | | DeepSeek-R1-Distill-Llama-70B | 70.0 | **86.7** | **94.5** | **65.2** | **57.5** | 1633 | </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` **NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output.** ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).

mradermacher/LwQ-Reasoner-10B-GGUF

mradermacher

2025-01-20T23:23:24Z

345

0

transformers

[ "transformers", "gguf", "LlamaWithQuestions", "CoT", "Reasoner", "LWQ", "en", "base_model:prithivMLmods/LwQ-Reasoner-10B", "base_model:quantized:prithivMLmods/LwQ-Reasoner-10B", "license:llama3.1", "endpoints_compatible", "region:us" ]

null

2025-01-20T16:50:16Z

--- base_model: prithivMLmods/LwQ-Reasoner-10B language: - en library_name: transformers license: llama3.1 quantized_by: mradermacher tags: - LlamaWithQuestions - CoT - Reasoner - LWQ --- ## About      static quants of https://huggingface.co/prithivMLmods/LwQ-Reasoner-10B  weighted/imatrix quants are available at https://huggingface.co/mradermacher/LwQ-Reasoner-10B-i1-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q2_K.gguf) | Q2_K | 4.0 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q3_K_S.gguf) | Q3_K_S | 4.7 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q3_K_M.gguf) | Q3_K_M | 5.2 | lower quality | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q3_K_L.gguf) | Q3_K_L | 5.6 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.IQ4_XS.gguf) | IQ4_XS | 5.8 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q4_K_S.gguf) | Q4_K_S | 6.1 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q4_K_M.gguf) | Q4_K_M | 6.4 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q5_K_S.gguf) | Q5_K_S | 7.2 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q5_K_M.gguf) | Q5_K_M | 7.4 | | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q6_K.gguf) | Q6_K | 8.6 | very good quality | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.Q8_0.gguf) | Q8_0 | 11.1 | fast, best quality | | [GGUF](https://huggingface.co/mradermacher/LwQ-Reasoner-10B-GGUF/resolve/main/LwQ-Reasoner-10B.f16.gguf) | f16 | 20.7 | 16 bpw, overkill | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time.

LHRuig/hollycine

LHRuig

2025-01-20T23:23:02Z

6

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:22:02Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # hollycine <Gallery /> ## Model description hollycine lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/hollycine/tree/main) them in the Files & versions tab.

adamo1139/DeepSeek-R1-Distill-Qwen-1.5B-5bpw-exl2

adamo1139

2025-01-20T23:22:53Z

5

0

null

[ "qwen2", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "5-bit", "exl2", "region:us" ]

null

2025-01-20T22:51:52Z

--- base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --- # DeepSeek-R1    <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- **Post-Training: Large-Scale Reinforcement Learning on the Base Model** - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- **Distillation: Smaller Models Can Be Powerful Too** - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> | **Model** | **#Total Params** | **#Activated Params** | **Context Length** | **Download** | | :------------: | :------------: | :------------: | :------------: | :------------: | | DeepSeek-R1-Zero | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) | | DeepSeek-R1 | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) | </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> | **Model** | **Base Model** | **Download** | | :------------: | :------------: | :------------: | | DeepSeek-R1-Distill-Qwen-1.5B | [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) | | DeepSeek-R1-Distill-Qwen-7B | [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) | | DeepSeek-R1-Distill-Llama-8B | [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) | | DeepSeek-R1-Distill-Qwen-14B | [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) | |DeepSeek-R1-Distill-Qwen-32B | [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) | | DeepSeek-R1-Distill-Llama-70B | [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) | </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> | Category | Benchmark (Metric) | Claude-3.5-Sonnet-1022 | GPT-4o 0513 | DeepSeek V3 | OpenAI o1-mini | OpenAI o1-1217 | DeepSeek R1 | |----------|-------------------|----------------------|------------|--------------|----------------|------------|--------------| | | Architecture | - | - | MoE | - | - | MoE | | | # Activated Params | - | - | 37B | - | - | 37B | | | # Total Params | - | - | 671B | - | - | 671B | | English | MMLU (Pass@1) | 88.3 | 87.2 | 88.5 | 85.2 | **91.8** | 90.8 | | | MMLU-Redux (EM) | 88.9 | 88.0 | 89.1 | 86.7 | - | **92.9** | | | MMLU-Pro (EM) | 78.0 | 72.6 | 75.9 | 80.3 | - | **84.0** | | | DROP (3-shot F1) | 88.3 | 83.7 | 91.6 | 83.9 | 90.2 | **92.2** | | | IF-Eval (Prompt Strict) | **86.5** | 84.3 | 86.1 | 84.8 | - | 83.3 | | | GPQA-Diamond (Pass@1) | 65.0 | 49.9 | 59.1 | 60.0 | **75.7** | 71.5 | | | SimpleQA (Correct) | 28.4 | 38.2 | 24.9 | 7.0 | **47.0** | 30.1 | | | FRAMES (Acc.) | 72.5 | 80.5 | 73.3 | 76.9 | - | **82.5** | | | AlpacaEval2.0 (LC-winrate) | 52.0 | 51.1 | 70.0 | 57.8 | - | **87.6** | | | ArenaHard (GPT-4-1106) | 85.2 | 80.4 | 85.5 | 92.0 | - | **92.3** | | Code | LiveCodeBench (Pass@1-COT) | 33.8 | 34.2 | - | 53.8 | 63.4 | **65.9** | | | Codeforces (Percentile) | 20.3 | 23.6 | 58.7 | 93.4 | **96.6** | 96.3 | | | Codeforces (Rating) | 717 | 759 | 1134 | 1820 | **2061** | 2029 | | | SWE Verified (Resolved) | **50.8** | 38.8 | 42.0 | 41.6 | 48.9 | 49.2 | | | Aider-Polyglot (Acc.) | 45.3 | 16.0 | 49.6 | 32.9 | **61.7** | 53.3 | | Math | AIME 2024 (Pass@1) | 16.0 | 9.3 | 39.2 | 63.6 | 79.2 | **79.8** | | | MATH-500 (Pass@1) | 78.3 | 74.6 | 90.2 | 90.0 | 96.4 | **97.3** | | | CNMO 2024 (Pass@1) | 13.1 | 10.8 | 43.2 | 67.6 | - | **78.8** | | Chinese | CLUEWSC (EM) | 85.4 | 87.9 | 90.9 | 89.9 | - | **92.8** | | | C-Eval (EM) | 76.7 | 76.0 | 86.5 | 68.9 | - | **91.8** | | | C-SimpleQA (Correct) | 55.4 | 58.7 | **68.0** | 40.3 | - | 63.7 | </div> ### Distilled Model Evaluation <div align="center"> | Model | AIME 2024 pass@1 | AIME 2024 cons@64 | MATH-500 pass@1 | GPQA Diamond pass@1 | LiveCodeBench pass@1 | CodeForces rating | |------------------------------------------|------------------|-------------------|-----------------|----------------------|----------------------|-------------------| | GPT-4o-0513 | 9.3 | 13.4 | 74.6 | 49.9 | 32.9 | 759 | | Claude-3.5-Sonnet-1022 | 16.0 | 26.7 | 78.3 | 65.0 | 38.9 | 717 | | o1-mini | 63.6 | 80.0 | 90.0 | 60.0 | 53.8 | **1820** | | QwQ-32B-Preview | 44.0 | 60.0 | 90.6 | 54.5 | 41.9 | 1316 | | DeepSeek-R1-Distill-Qwen-1.5B | 28.9 | 52.7 | 83.9 | 33.8 | 16.9 | 954 | | DeepSeek-R1-Distill-Qwen-7B | 55.5 | 83.3 | 92.8 | 49.1 | 37.6 | 1189 | | DeepSeek-R1-Distill-Qwen-14B | 69.7 | 80.0 | 93.9 | 59.1 | 53.1 | 1481 | | DeepSeek-R1-Distill-Qwen-32B | **72.6** | 83.3 | 94.3 | 62.1 | 57.2 | 1691 | | DeepSeek-R1-Distill-Llama-8B | 50.4 | 80.0 | 89.1 | 49.0 | 39.6 | 1205 | | DeepSeek-R1-Distill-Llama-70B | 70.0 | **86.7** | **94.5** | **65.2** | **57.5** | 1633 | </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` **NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output.** ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).

oldiday/625b4c61-4c53-4c35-8346-7973e6e5d4d4

oldiday

2025-01-20T23:22:12Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2.5-Coder-7B-Instruct", "base_model:adapter:unsloth/Qwen2.5-Coder-7B-Instruct", "license:apache-2.0", "region:us" ]

null

2025-01-20T22:55:44Z

--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2.5-Coder-7B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 625b4c61-4c53-4c35-8346-7973e6e5d4d4 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2.5-Coder-7B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 5da4fdb4f9d40cf6_train_data.json ds_type: json format: custom path: /workspace/input_data/5da4fdb4f9d40cf6_train_data.json type: field_input: topic; field_instruction: message_1 field_output: message_2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: oldiday/625b4c61-4c53-4c35-8346-7973e6e5d4d4 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 5.0e-05 load_in_4bit: false load_in_8bit: false local_rank: 0 logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 100 micro_batch_size: 8 mlflow_experiment_name: /tmp/5da4fdb4f9d40cf6_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: techspear-hub wandb_mode: online wandb_name: 971c59eb-5de8-4a78-8d22-6a7da4c9ee82 wandb_project: Gradients-On-Six wandb_run: your_name wandb_runid: 971c59eb-5de8-4a78-8d22-6a7da4c9ee82 warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # 625b4c61-4c53-4c35-8346-7973e6e5d4d4 This model is a fine-tuned version of [unsloth/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.5959 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0017 | 1 | 0.6948 | | 0.6968 | 0.0153 | 9 | 0.6832 | | 0.6562 | 0.0306 | 18 | 0.6471 | | 0.6206 | 0.0459 | 27 | 0.6231 | | 0.6178 | 0.0612 | 36 | 0.6112 | | 0.625 | 0.0765 | 45 | 0.6054 | | 0.605 | 0.0918 | 54 | 0.6015 | | 0.5935 | 0.1071 | 63 | 0.5990 | | 0.6049 | 0.1224 | 72 | 0.5973 | | 0.5989 | 0.1378 | 81 | 0.5964 | | 0.5937 | 0.1531 | 90 | 0.5960 | | 0.6079 | 0.1684 | 99 | 0.5959 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

MayBashendy/ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k18_task5_organization

MayBashendy

2025-01-20T23:22:05Z

5

0

transformers

[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]

text-classification

2025-01-20T23:10:20Z

--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k18_task5_organization results: [] ---  # ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k18_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2288 - Qwk: 0.0 - Mse: 1.2288 - Rmse: 1.1085 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | Qwk | Mse | Rmse | |:-------------:|:-------:|:----:|:---------------:|:-------:|:------:|:------:| | No log | 0.0465 | 2 | 4.0158 | 0.0034 | 4.0158 | 2.0039 | | No log | 0.0930 | 4 | 2.2748 | 0.0271 | 2.2748 | 1.5083 | | No log | 0.1395 | 6 | 1.6688 | 0.0329 | 1.6688 | 1.2918 | | No log | 0.1860 | 8 | 1.2346 | 0.0967 | 1.2346 | 1.1111 | | No log | 0.2326 | 10 | 1.1925 | 0.0731 | 1.1925 | 1.0920 | | No log | 0.2791 | 12 | 1.1109 | 0.2150 | 1.1109 | 1.0540 | | No log | 0.3256 | 14 | 1.0932 | 0.1884 | 1.0932 | 1.0456 | | No log | 0.3721 | 16 | 1.2148 | 0.0380 | 1.2148 | 1.1022 | | No log | 0.4186 | 18 | 1.2750 | 0.0 | 1.2750 | 1.1292 | | No log | 0.4651 | 20 | 1.1496 | 0.0760 | 1.1496 | 1.0722 | | No log | 0.5116 | 22 | 1.1357 | 0.1379 | 1.1357 | 1.0657 | | No log | 0.5581 | 24 | 1.1803 | 0.1333 | 1.1803 | 1.0864 | | No log | 0.6047 | 26 | 1.1167 | 0.1981 | 1.1167 | 1.0568 | | No log | 0.6512 | 28 | 1.1559 | 0.0445 | 1.1559 | 1.0751 | | No log | 0.6977 | 30 | 1.2187 | 0.0445 | 1.2187 | 1.1039 | | No log | 0.7442 | 32 | 1.2232 | 0.0445 | 1.2232 | 1.1060 | | No log | 0.7907 | 34 | 1.1709 | 0.1643 | 1.1709 | 1.0821 | | No log | 0.8372 | 36 | 1.1800 | 0.0792 | 1.1800 | 1.0863 | | No log | 0.8837 | 38 | 1.3732 | -0.0296 | 1.3732 | 1.1719 | | No log | 0.9302 | 40 | 1.6117 | 0.0143 | 1.6117 | 1.2695 | | No log | 0.9767 | 42 | 1.7119 | 0.0143 | 1.7119 | 1.3084 | | No log | 1.0233 | 44 | 1.5533 | 0.0 | 1.5533 | 1.2463 | | No log | 1.0698 | 46 | 1.4371 | -0.0148 | 1.4371 | 1.1988 | | No log | 1.1163 | 48 | 1.1990 | 0.0701 | 1.1990 | 1.0950 | | No log | 1.1628 | 50 | 1.0921 | 0.2239 | 1.0921 | 1.0450 | | No log | 1.2093 | 52 | 1.2033 | 0.0102 | 1.2033 | 1.0969 | | No log | 1.2558 | 54 | 1.2896 | -0.1560 | 1.2896 | 1.1356 | | No log | 1.3023 | 56 | 1.4626 | -0.0326 | 1.4626 | 1.2094 | | No log | 1.3488 | 58 | 1.3772 | -0.0022 | 1.3772 | 1.1735 | | No log | 1.3953 | 60 | 1.2187 | 0.1361 | 1.2187 | 1.1040 | | No log | 1.4419 | 62 | 1.2426 | 0.0374 | 1.2426 | 1.1147 | | No log | 1.4884 | 64 | 1.2838 | 0.0999 | 1.2838 | 1.1330 | | No log | 1.5349 | 66 | 1.2161 | 0.0700 | 1.2161 | 1.1028 | | No log | 1.5814 | 68 | 1.1278 | 0.2692 | 1.1278 | 1.0620 | | No log | 1.6279 | 70 | 1.1283 | 0.1725 | 1.1283 | 1.0622 | | No log | 1.6744 | 72 | 1.0934 | 0.2366 | 1.0934 | 1.0456 | | No log | 1.7209 | 74 | 1.0742 | 0.1783 | 1.0742 | 1.0364 | | No log | 1.7674 | 76 | 1.1562 | 0.1389 | 1.1562 | 1.0752 | | No log | 1.8140 | 78 | 1.4371 | -0.0641 | 1.4371 | 1.1988 | | No log | 1.8605 | 80 | 1.4812 | -0.0641 | 1.4812 | 1.2171 | | No log | 1.9070 | 82 | 1.4368 | -0.0641 | 1.4368 | 1.1987 | | No log | 1.9535 | 84 | 1.2691 | -0.0091 | 1.2691 | 1.1266 | | No log | 2.0 | 86 | 1.2183 | 0.0542 | 1.2183 | 1.1038 | | No log | 2.0465 | 88 | 1.2777 | 0.0188 | 1.2777 | 1.1304 | | No log | 2.0930 | 90 | 1.4962 | -0.1067 | 1.4962 | 1.2232 | | No log | 2.1395 | 92 | 1.5524 | -0.0747 | 1.5524 | 1.2460 | | No log | 2.1860 | 94 | 1.3420 | 0.0587 | 1.3420 | 1.1585 | | No log | 2.2326 | 96 | 1.1766 | 0.0164 | 1.1766 | 1.0847 | | No log | 2.2791 | 98 | 1.1748 | 0.0164 | 1.1748 | 1.0839 | | No log | 2.3256 | 100 | 1.3960 | 0.0464 | 1.3960 | 1.1815 | | No log | 2.3721 | 102 | 1.7947 | -0.1843 | 1.7947 | 1.3397 | | No log | 2.4186 | 104 | 1.9849 | -0.3093 | 1.9849 | 1.4089 | | No log | 2.4651 | 106 | 1.9655 | -0.1443 | 1.9655 | 1.4020 | | No log | 2.5116 | 108 | 1.7718 | -0.0788 | 1.7718 | 1.3311 | | No log | 2.5581 | 110 | 1.5684 | -0.0167 | 1.5684 | 1.2524 | | No log | 2.6047 | 112 | 1.6681 | -0.0688 | 1.6681 | 1.2916 | | No log | 2.6512 | 114 | 1.8768 | -0.1179 | 1.8768 | 1.3700 | | No log | 2.6977 | 116 | 2.0265 | -0.1122 | 2.0265 | 1.4236 | | No log | 2.7442 | 118 | 2.0327 | -0.1111 | 2.0327 | 1.4257 | | No log | 2.7907 | 120 | 2.0567 | -0.1154 | 2.0567 | 1.4341 | | No log | 2.8372 | 122 | 1.8254 | -0.0655 | 1.8254 | 1.3511 | | No log | 2.8837 | 124 | 1.5249 | -0.0735 | 1.5249 | 1.2349 | | No log | 2.9302 | 126 | 1.7057 | -0.1078 | 1.7057 | 1.3060 | | No log | 2.9767 | 128 | 2.0800 | -0.2156 | 2.0800 | 1.4422 | | No log | 3.0233 | 130 | 2.3606 | -0.2661 | 2.3606 | 1.5364 | | No log | 3.0698 | 132 | 2.2816 | -0.1571 | 2.2816 | 1.5105 | | No log | 3.1163 | 134 | 2.0213 | -0.1658 | 2.0213 | 1.4217 | | No log | 3.1628 | 136 | 1.9379 | -0.1254 | 1.9379 | 1.3921 | | No log | 3.2093 | 138 | 1.8792 | -0.0892 | 1.8792 | 1.3708 | | No log | 3.2558 | 140 | 1.8788 | -0.0849 | 1.8788 | 1.3707 | | No log | 3.3023 | 142 | 1.7651 | -0.0762 | 1.7651 | 1.3286 | | No log | 3.3488 | 144 | 1.5874 | 0.0294 | 1.5874 | 1.2599 | | No log | 3.3953 | 146 | 1.5857 | 0.0294 | 1.5857 | 1.2593 | | No log | 3.4419 | 148 | 1.6411 | 0.0279 | 1.6411 | 1.2810 | | No log | 3.4884 | 150 | 1.7297 | -0.0267 | 1.7297 | 1.3152 | | No log | 3.5349 | 152 | 1.7574 | -0.0757 | 1.7574 | 1.3257 | | No log | 3.5814 | 154 | 1.7628 | -0.0397 | 1.7628 | 1.3277 | | No log | 3.6279 | 156 | 1.7589 | -0.0508 | 1.7589 | 1.3262 | | No log | 3.6744 | 158 | 1.7224 | -0.0378 | 1.7224 | 1.3124 | | No log | 3.7209 | 160 | 1.7568 | -0.0514 | 1.7568 | 1.3254 | | No log | 3.7674 | 162 | 1.8652 | 0.0726 | 1.8652 | 1.3657 | | No log | 3.8140 | 164 | 1.9175 | 0.0402 | 1.9175 | 1.3847 | | No log | 3.8605 | 166 | 1.8288 | 0.0400 | 1.8288 | 1.3523 | | No log | 3.9070 | 168 | 1.8386 | 0.0562 | 1.8386 | 1.3559 | | No log | 3.9535 | 170 | 1.8316 | 0.1342 | 1.8316 | 1.3534 | | No log | 4.0 | 172 | 1.6490 | 0.2105 | 1.6490 | 1.2842 | | No log | 4.0465 | 174 | 1.5040 | 0.1282 | 1.5040 | 1.2264 | | No log | 4.0930 | 176 | 1.4472 | 0.1198 | 1.4472 | 1.2030 | | No log | 4.1395 | 178 | 1.4452 | 0.0270 | 1.4452 | 1.2022 | | No log | 4.1860 | 180 | 1.5266 | 0.0809 | 1.5266 | 1.2356 | | No log | 4.2326 | 182 | 1.7969 | 0.1525 | 1.7969 | 1.3405 | | No log | 4.2791 | 184 | 1.8888 | 0.2098 | 1.8888 | 1.3744 | | No log | 4.3256 | 186 | 1.7524 | 0.1663 | 1.7524 | 1.3238 | | No log | 4.3721 | 188 | 1.5118 | 0.2342 | 1.5118 | 1.2296 | | No log | 4.4186 | 190 | 1.4063 | 0.0946 | 1.4063 | 1.1859 | | No log | 4.4651 | 192 | 1.4720 | 0.1911 | 1.4720 | 1.2133 | | No log | 4.5116 | 194 | 1.4943 | 0.1423 | 1.4943 | 1.2224 | | No log | 4.5581 | 196 | 1.4577 | 0.1423 | 1.4577 | 1.2073 | | No log | 4.6047 | 198 | 1.3969 | 0.0602 | 1.3969 | 1.1819 | | No log | 4.6512 | 200 | 1.3910 | 0.1110 | 1.3910 | 1.1794 | | No log | 4.6977 | 202 | 1.4709 | 0.1110 | 1.4709 | 1.2128 | | No log | 4.7442 | 204 | 1.5518 | 0.1423 | 1.5518 | 1.2457 | | No log | 4.7907 | 206 | 1.5587 | 0.1027 | 1.5587 | 1.2485 | | No log | 4.8372 | 208 | 1.5567 | 0.1027 | 1.5567 | 1.2477 | | No log | 4.8837 | 210 | 1.5407 | 0.1423 | 1.5407 | 1.2413 | | No log | 4.9302 | 212 | 1.4965 | 0.1904 | 1.4965 | 1.2233 | | No log | 4.9767 | 214 | 1.3997 | 0.1943 | 1.3997 | 1.1831 | | No log | 5.0233 | 216 | 1.3739 | 0.2126 | 1.3739 | 1.1721 | | No log | 5.0698 | 218 | 1.4697 | 0.2292 | 1.4697 | 1.2123 | | No log | 5.1163 | 220 | 1.4313 | 0.2015 | 1.4313 | 1.1964 | | No log | 5.1628 | 222 | 1.4067 | 0.2203 | 1.4067 | 1.1860 | | No log | 5.2093 | 224 | 1.4171 | 0.1886 | 1.4171 | 1.1904 | | No log | 5.2558 | 226 | 1.5312 | 0.2117 | 1.5312 | 1.2374 | | No log | 5.3023 | 228 | 1.7132 | 0.2058 | 1.7132 | 1.3089 | | No log | 5.3488 | 230 | 1.7432 | 0.2206 | 1.7432 | 1.3203 | | No log | 5.3953 | 232 | 1.7312 | 0.1963 | 1.7312 | 1.3158 | | No log | 5.4419 | 234 | 1.7061 | 0.2389 | 1.7061 | 1.3062 | | No log | 5.4884 | 236 | 1.5999 | 0.1461 | 1.5999 | 1.2649 | | No log | 5.5349 | 238 | 1.5428 | 0.1911 | 1.5428 | 1.2421 | | No log | 5.5814 | 240 | 1.3937 | 0.1142 | 1.3937 | 1.1806 | | No log | 5.6279 | 242 | 1.3014 | 0.1052 | 1.3014 | 1.1408 | | No log | 5.6744 | 244 | 1.2485 | 0.1052 | 1.2485 | 1.1174 | | No log | 5.7209 | 246 | 1.2400 | 0.0401 | 1.2400 | 1.1136 | | No log | 5.7674 | 248 | 1.2161 | 0.0401 | 1.2161 | 1.1028 | | No log | 5.8140 | 250 | 1.1670 | 0.0556 | 1.1670 | 1.0803 | | No log | 5.8605 | 252 | 1.1944 | 0.0155 | 1.1944 | 1.0929 | | No log | 5.9070 | 254 | 1.3146 | 0.0781 | 1.3146 | 1.1466 | | No log | 5.9535 | 256 | 1.5336 | 0.1880 | 1.5336 | 1.2384 | | No log | 6.0 | 258 | 1.6128 | 0.2465 | 1.6128 | 1.2700 | | No log | 6.0465 | 260 | 1.5854 | 0.2465 | 1.5854 | 1.2591 | | No log | 6.0930 | 262 | 1.5079 | 0.1966 | 1.5079 | 1.2280 | | No log | 6.1395 | 264 | 1.4465 | 0.1886 | 1.4465 | 1.2027 | | No log | 6.1860 | 266 | 1.4232 | 0.2027 | 1.4232 | 1.1930 | | No log | 6.2326 | 268 | 1.3952 | 0.2027 | 1.3952 | 1.1812 | | No log | 6.2791 | 270 | 1.3799 | 0.2027 | 1.3799 | 1.1747 | | No log | 6.3256 | 272 | 1.3777 | 0.2089 | 1.3777 | 1.1738 | | No log | 6.3721 | 274 | 1.4345 | 0.2203 | 1.4345 | 1.1977 | | No log | 6.4186 | 276 | 1.4906 | 0.2555 | 1.4906 | 1.2209 | | No log | 6.4651 | 278 | 1.6231 | 0.2771 | 1.6231 | 1.2740 | | No log | 6.5116 | 280 | 1.6548 | 0.2270 | 1.6548 | 1.2864 | | No log | 6.5581 | 282 | 1.4900 | 0.1058 | 1.4900 | 1.2207 | | No log | 6.6047 | 284 | 1.4021 | 0.0781 | 1.4021 | 1.1841 | | No log | 6.6512 | 286 | 1.4033 | 0.0781 | 1.4033 | 1.1846 | | No log | 6.6977 | 288 | 1.4786 | 0.2315 | 1.4786 | 1.2160 | | No log | 6.7442 | 290 | 1.4991 | 0.2915 | 1.4991 | 1.2244 | | No log | 6.7907 | 292 | 1.4009 | 0.2455 | 1.4009 | 1.1836 | | No log | 6.8372 | 294 | 1.3681 | 0.2455 | 1.3681 | 1.1697 | | No log | 6.8837 | 296 | 1.3522 | 0.2455 | 1.3522 | 1.1628 | | No log | 6.9302 | 298 | 1.3574 | 0.2506 | 1.3574 | 1.1651 | | No log | 6.9767 | 300 | 1.4005 | 0.2506 | 1.4005 | 1.1834 | | No log | 7.0233 | 302 | 1.4022 | 0.2203 | 1.4022 | 1.1841 | | No log | 7.0698 | 304 | 1.3696 | 0.1886 | 1.3696 | 1.1703 | | No log | 7.1163 | 306 | 1.3437 | 0.1886 | 1.3437 | 1.1592 | | No log | 7.1628 | 308 | 1.3611 | 0.1886 | 1.3611 | 1.1667 | | No log | 7.2093 | 310 | 1.4336 | 0.2260 | 1.4336 | 1.1973 | | No log | 7.2558 | 312 | 1.4199 | 0.1886 | 1.4199 | 1.1916 | | No log | 7.3023 | 314 | 1.3951 | 0.1886 | 1.3951 | 1.1811 | | No log | 7.3488 | 316 | 1.3180 | 0.1316 | 1.3180 | 1.1481 | | No log | 7.3953 | 318 | 1.3180 | 0.1548 | 1.3180 | 1.1480 | | No log | 7.4419 | 320 | 1.4264 | 0.1980 | 1.4264 | 1.1943 | | No log | 7.4884 | 322 | 1.7047 | 0.2315 | 1.7047 | 1.3057 | | No log | 7.5349 | 324 | 1.7725 | 0.2116 | 1.7725 | 1.3313 | | No log | 7.5814 | 326 | 1.6272 | 0.2363 | 1.6272 | 1.2756 | | No log | 7.6279 | 328 | 1.5210 | 0.2395 | 1.5210 | 1.2333 | | No log | 7.6744 | 330 | 1.4116 | 0.2027 | 1.4116 | 1.1881 | | No log | 7.7209 | 332 | 1.3019 | 0.1622 | 1.3019 | 1.1410 | | No log | 7.7674 | 334 | 1.2722 | 0.1622 | 1.2722 | 1.1279 | | No log | 7.8140 | 336 | 1.3118 | 0.1473 | 1.3118 | 1.1453 | | No log | 7.8605 | 338 | 1.3846 | 0.1886 | 1.3846 | 1.1767 | | No log | 7.9070 | 340 | 1.3764 | 0.1886 | 1.3764 | 1.1732 | | No log | 7.9535 | 342 | 1.3974 | 0.1886 | 1.3974 | 1.1821 | | No log | 8.0 | 344 | 1.4388 | 0.2506 | 1.4388 | 1.1995 | | No log | 8.0465 | 346 | 1.4310 | 0.2126 | 1.4310 | 1.1962 | | No log | 8.0930 | 348 | 1.4688 | 0.2424 | 1.4688 | 1.2119 | | No log | 8.1395 | 350 | 1.4365 | 0.2424 | 1.4365 | 1.1986 | | No log | 8.1860 | 352 | 1.3914 | 0.1814 | 1.3914 | 1.1796 | | No log | 8.2326 | 354 | 1.3864 | 0.1814 | 1.3864 | 1.1775 | | No log | 8.2791 | 356 | 1.3512 | 0.1552 | 1.3512 | 1.1624 | | No log | 8.3256 | 358 | 1.3229 | 0.1473 | 1.3229 | 1.1502 | | No log | 8.3721 | 360 | 1.3194 | 0.1552 | 1.3194 | 1.1486 | | No log | 8.4186 | 362 | 1.3151 | 0.1552 | 1.3151 | 1.1468 | | No log | 8.4651 | 364 | 1.4077 | 0.2640 | 1.4077 | 1.1865 | | No log | 8.5116 | 366 | 1.5515 | 0.2223 | 1.5515 | 1.2456 | | No log | 8.5581 | 368 | 1.6437 | 0.1978 | 1.6437 | 1.2821 | | No log | 8.6047 | 370 | 1.7347 | 0.2414 | 1.7347 | 1.3171 | | No log | 8.6512 | 372 | 1.8221 | 0.2224 | 1.8221 | 1.3499 | | No log | 8.6977 | 374 | 1.6633 | 0.2481 | 1.6633 | 1.2897 | | No log | 8.7442 | 376 | 1.4077 | 0.1898 | 1.4077 | 1.1865 | | No log | 8.7907 | 378 | 1.3061 | 0.0931 | 1.3061 | 1.1428 | | No log | 8.8372 | 380 | 1.2222 | 0.0160 | 1.2222 | 1.1055 | | No log | 8.8837 | 382 | 1.2155 | 0.0160 | 1.2155 | 1.1025 | | No log | 8.9302 | 384 | 1.2623 | 0.0445 | 1.2623 | 1.1235 | | No log | 8.9767 | 386 | 1.3885 | 0.1552 | 1.3885 | 1.1783 | | No log | 9.0233 | 388 | 1.4957 | 0.2795 | 1.4957 | 1.2230 | | No log | 9.0698 | 390 | 1.3858 | 0.2640 | 1.3858 | 1.1772 | | No log | 9.1163 | 392 | 1.2752 | 0.1961 | 1.2752 | 1.1293 | | No log | 9.1628 | 394 | 1.2226 | 0.1961 | 1.2226 | 1.1057 | | No log | 9.2093 | 396 | 1.1909 | 0.1697 | 1.1909 | 1.0913 | | No log | 9.2558 | 398 | 1.1800 | 0.1552 | 1.1800 | 1.0863 | | No log | 9.3023 | 400 | 1.1418 | 0.1202 | 1.1418 | 1.0686 | | No log | 9.3488 | 402 | 1.1663 | 0.1552 | 1.1663 | 1.0800 | | No log | 9.3953 | 404 | 1.1681 | 0.1552 | 1.1681 | 1.0808 | | No log | 9.4419 | 406 | 1.2341 | 0.1552 | 1.2341 | 1.1109 | | No log | 9.4884 | 408 | 1.3670 | 0.2647 | 1.3670 | 1.1692 | | No log | 9.5349 | 410 | 1.4327 | 0.2465 | 1.4327 | 1.1970 | | No log | 9.5814 | 412 | 1.4405 | 0.2465 | 1.4405 | 1.2002 | | No log | 9.6279 | 414 | 1.4134 | 0.2602 | 1.4134 | 1.1889 | | No log | 9.6744 | 416 | 1.3585 | 0.2506 | 1.3585 | 1.1656 | | No log | 9.7209 | 418 | 1.3508 | 0.1886 | 1.3508 | 1.1623 | | No log | 9.7674 | 420 | 1.3591 | 0.1952 | 1.3591 | 1.1658 | | No log | 9.8140 | 422 | 1.3131 | 0.1952 | 1.3131 | 1.1459 | | No log | 9.8605 | 424 | 1.3215 | 0.1886 | 1.3215 | 1.1496 | | No log | 9.9070 | 426 | 1.4130 | 0.2506 | 1.4130 | 1.1887 | | No log | 9.9535 | 428 | 1.3922 | 0.1310 | 1.3922 | 1.1799 | | No log | 10.0 | 430 | 1.4431 | 0.1428 | 1.4431 | 1.2013 | | No log | 10.0465 | 432 | 1.5592 | 0.2117 | 1.5592 | 1.2487 | | No log | 10.0930 | 434 | 1.5122 | 0.2292 | 1.5122 | 1.2297 | | No log | 10.1395 | 436 | 1.4766 | 0.2647 | 1.4766 | 1.2151 | | No log | 10.1860 | 438 | 1.3397 | 0.2315 | 1.3397 | 1.1574 | | No log | 10.2326 | 440 | 1.2819 | 0.1886 | 1.2819 | 1.1322 | | No log | 10.2791 | 442 | 1.2954 | 0.1700 | 1.2954 | 1.1381 | | No log | 10.3256 | 444 | 1.3657 | 0.2315 | 1.3657 | 1.1686 | | No log | 10.3721 | 446 | 1.4421 | 0.2315 | 1.4421 | 1.2009 | | No log | 10.4186 | 448 | 1.4492 | 0.2015 | 1.4492 | 1.2038 | | No log | 10.4651 | 450 | 1.3858 | 0.1202 | 1.3858 | 1.1772 | | No log | 10.5116 | 452 | 1.3957 | 0.0833 | 1.3957 | 1.1814 | | No log | 10.5581 | 454 | 1.4324 | 0.0833 | 1.4324 | 1.1968 | | No log | 10.6047 | 456 | 1.5332 | 0.1552 | 1.5332 | 1.2382 | | No log | 10.6512 | 458 | 1.5961 | 0.1952 | 1.5961 | 1.2634 | | No log | 10.6977 | 460 | 1.6255 | 0.1595 | 1.6255 | 1.2750 | | No log | 10.7442 | 462 | 1.5482 | 0.1634 | 1.5482 | 1.2443 | | No log | 10.7907 | 464 | 1.4990 | 0.1407 | 1.4990 | 1.2243 | | No log | 10.8372 | 466 | 1.4643 | 0.1473 | 1.4643 | 1.2101 | | No log | 10.8837 | 468 | 1.4725 | 0.1407 | 1.4725 | 1.2135 | | No log | 10.9302 | 470 | 1.4109 | 0.0781 | 1.4109 | 1.1878 | | No log | 10.9767 | 472 | 1.3841 | 0.1142 | 1.3841 | 1.1765 | | No log | 11.0233 | 474 | 1.4789 | 0.2239 | 1.4789 | 1.2161 | | No log | 11.0698 | 476 | 1.5078 | 0.2126 | 1.5078 | 1.2279 | | No log | 11.1163 | 478 | 1.5220 | 0.1486 | 1.5220 | 1.2337 | | No log | 11.1628 | 480 | 1.4774 | 0.1552 | 1.4774 | 1.2155 | | No log | 11.2093 | 482 | 1.4310 | 0.1486 | 1.4310 | 1.1962 | | No log | 11.2558 | 484 | 1.4349 | 0.2424 | 1.4349 | 1.1979 | | No log | 11.3023 | 486 | 1.3930 | 0.2424 | 1.3930 | 1.1803 | | No log | 11.3488 | 488 | 1.3830 | 0.2126 | 1.3830 | 1.1760 | | No log | 11.3953 | 490 | 1.3882 | 0.1407 | 1.3882 | 1.1782 | | No log | 11.4419 | 492 | 1.3933 | 0.1473 | 1.3933 | 1.1804 | | No log | 11.4884 | 494 | 1.3616 | 0.1473 | 1.3616 | 1.1669 | | No log | 11.5349 | 496 | 1.3209 | 0.1473 | 1.3209 | 1.1493 | | No log | 11.5814 | 498 | 1.2735 | 0.0401 | 1.2735 | 1.1285 | | 0.3529 | 11.6279 | 500 | 1.2606 | 0.0401 | 1.2606 | 1.1228 | | 0.3529 | 11.6744 | 502 | 1.2457 | 0.0401 | 1.2457 | 1.1161 | | 0.3529 | 11.7209 | 504 | 1.2800 | 0.0401 | 1.2800 | 1.1314 | | 0.3529 | 11.7674 | 506 | 1.3705 | 0.1052 | 1.3705 | 1.1707 | | 0.3529 | 11.8140 | 508 | 1.4245 | 0.1351 | 1.4245 | 1.1935 | | 0.3529 | 11.8605 | 510 | 1.4236 | 0.1351 | 1.4236 | 1.1932 | | 0.3529 | 11.9070 | 512 | 1.4686 | 0.1142 | 1.4686 | 1.2118 | | 0.3529 | 11.9535 | 514 | 1.5206 | 0.2126 | 1.5206 | 1.2331 | | 0.3529 | 12.0 | 516 | 1.4949 | 0.2126 | 1.4949 | 1.2226 | | 0.3529 | 12.0465 | 518 | 1.4517 | 0.1562 | 1.4517 | 1.2048 | | 0.3529 | 12.0930 | 520 | 1.4807 | 0.2126 | 1.4807 | 1.2169 | | 0.3529 | 12.1395 | 522 | 1.5059 | 0.2239 | 1.5059 | 1.2272 | | 0.3529 | 12.1860 | 524 | 1.6043 | 0.2391 | 1.6043 | 1.2666 | | 0.3529 | 12.2326 | 526 | 1.5695 | 0.2391 | 1.5695 | 1.2528 | | 0.3529 | 12.2791 | 528 | 1.4159 | 0.2424 | 1.4159 | 1.1899 | | 0.3529 | 12.3256 | 530 | 1.3219 | 0.1486 | 1.3219 | 1.1497 | | 0.3529 | 12.3721 | 532 | 1.3401 | 0.2065 | 1.3401 | 1.1576 | | 0.3529 | 12.4186 | 534 | 1.3882 | 0.2424 | 1.3882 | 1.1782 | | 0.3529 | 12.4651 | 536 | 1.3911 | 0.2126 | 1.3911 | 1.1795 | | 0.3529 | 12.5116 | 538 | 1.3736 | 0.2506 | 1.3736 | 1.1720 | | 0.3529 | 12.5581 | 540 | 1.3963 | 0.2424 | 1.3963 | 1.1816 | | 0.3529 | 12.6047 | 542 | 1.4377 | 0.2709 | 1.4377 | 1.1990 | | 0.3529 | 12.6512 | 544 | 1.3851 | 0.2126 | 1.3851 | 1.1769 | | 0.3529 | 12.6977 | 546 | 1.3814 | 0.2126 | 1.3814 | 1.1753 | | 0.3529 | 12.7442 | 548 | 1.4161 | 0.2424 | 1.4161 | 1.1900 | | 0.3529 | 12.7907 | 550 | 1.4834 | 0.2690 | 1.4834 | 1.2180 | | 0.3529 | 12.8372 | 552 | 1.5133 | 0.2731 | 1.5133 | 1.2302 | | 0.3529 | 12.8837 | 554 | 1.4658 | 0.2690 | 1.4658 | 1.2107 | | 0.3529 | 12.9302 | 556 | 1.3921 | 0.2315 | 1.3921 | 1.1799 | | 0.3529 | 12.9767 | 558 | 1.3701 | 0.2506 | 1.3701 | 1.1705 | | 0.3529 | 13.0233 | 560 | 1.4052 | 0.2795 | 1.4052 | 1.1854 | | 0.3529 | 13.0698 | 562 | 1.4930 | 0.2602 | 1.4930 | 1.2219 | | 0.3529 | 13.1163 | 564 | 1.5950 | 0.2731 | 1.5950 | 1.2629 | | 0.3529 | 13.1628 | 566 | 1.6690 | 0.3232 | 1.6690 | 1.2919 | | 0.3529 | 13.2093 | 568 | 1.6193 | 0.3172 | 1.6193 | 1.2725 | | 0.3529 | 13.2558 | 570 | 1.5363 | 0.2752 | 1.5363 | 1.2395 | | 0.3529 | 13.3023 | 572 | 1.4229 | 0.0833 | 1.4229 | 1.1928 | | 0.3529 | 13.3488 | 574 | 1.3506 | 0.0833 | 1.3506 | 1.1622 | | 0.3529 | 13.3953 | 576 | 1.3492 | 0.0833 | 1.3492 | 1.1616 | | 0.3529 | 13.4419 | 578 | 1.3896 | 0.0833 | 1.3896 | 1.1788 | | 0.3529 | 13.4884 | 580 | 1.4283 | 0.0781 | 1.4283 | 1.1951 | | 0.3529 | 13.5349 | 582 | 1.4724 | 0.1142 | 1.4724 | 1.2134 | | 0.3529 | 13.5814 | 584 | 1.4633 | 0.1142 | 1.4633 | 1.2097 | | 0.3529 | 13.6279 | 586 | 1.4614 | 0.1142 | 1.4614 | 1.2089 | | 0.3529 | 13.6744 | 588 | 1.4658 | 0.1486 | 1.4658 | 1.2107 | | 0.3529 | 13.7209 | 590 | 1.4604 | 0.1228 | 1.4604 | 1.2085 | | 0.3529 | 13.7674 | 592 | 1.3979 | 0.1228 | 1.3979 | 1.1823 | | 0.3529 | 13.8140 | 594 | 1.3791 | 0.1228 | 1.3791 | 1.1744 | | 0.3529 | 13.8605 | 596 | 1.4188 | 0.2126 | 1.4188 | 1.1911 | | 0.3529 | 13.9070 | 598 | 1.4513 | 0.2709 | 1.4513 | 1.2047 | | 0.3529 | 13.9535 | 600 | 1.4620 | 0.2709 | 1.4620 | 1.2092 | | 0.3529 | 14.0 | 602 | 1.3948 | 0.2424 | 1.3948 | 1.1810 | | 0.3529 | 14.0465 | 604 | 1.3774 | 0.2126 | 1.3774 | 1.1736 | | 0.3529 | 14.0930 | 606 | 1.3843 | 0.2424 | 1.3843 | 1.1765 | | 0.3529 | 14.1395 | 608 | 1.3777 | 0.1814 | 1.3777 | 1.1738 | | 0.3529 | 14.1860 | 610 | 1.3095 | 0.0781 | 1.3095 | 1.1443 | | 0.3529 | 14.2326 | 612 | 1.2472 | 0.0401 | 1.2472 | 1.1168 | | 0.3529 | 14.2791 | 614 | 1.2122 | 0.0401 | 1.2122 | 1.1010 | | 0.3529 | 14.3256 | 616 | 1.1979 | 0.0 | 1.1979 | 1.0945 | | 0.3529 | 14.3721 | 618 | 1.2415 | 0.0401 | 1.2415 | 1.1142 | | 0.3529 | 14.4186 | 620 | 1.2917 | 0.0781 | 1.2917 | 1.1365 | | 0.3529 | 14.4651 | 622 | 1.3478 | 0.1310 | 1.3478 | 1.1609 | | 0.3529 | 14.5116 | 624 | 1.4031 | 0.1634 | 1.4031 | 1.1845 | | 0.3529 | 14.5581 | 626 | 1.4545 | 0.2522 | 1.4545 | 1.2060 | | 0.3529 | 14.6047 | 628 | 1.4968 | 0.2522 | 1.4968 | 1.2234 | | 0.3529 | 14.6512 | 630 | 1.5350 | 0.2522 | 1.5350 | 1.2389 | | 0.3529 | 14.6977 | 632 | 1.5023 | 0.2239 | 1.5023 | 1.2257 | | 0.3529 | 14.7442 | 634 | 1.3888 | 0.0401 | 1.3888 | 1.1785 | | 0.3529 | 14.7907 | 636 | 1.2818 | 0.0 | 1.2818 | 1.1322 | | 0.3529 | 14.8372 | 638 | 1.2042 | 0.0 | 1.2042 | 1.0974 | | 0.3529 | 14.8837 | 640 | 1.1860 | 0.0 | 1.1860 | 1.0890 | | 0.3529 | 14.9302 | 642 | 1.2288 | 0.0 | 1.2288 | 1.1085 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1

datlaaaaaaa/e8cccbe1-c136-409a-a725-9f09127c1a3f

datlaaaaaaa

2025-01-20T23:21:58Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:54:13Z

--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: e8cccbe1-c136-409a-a725-9f09127c1a3f results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 854bca96bed40197_train_data.json ds_type: json format: custom path: /workspace/input_data/854bca96bed40197_train_data.json type: field_input: state_before field_instruction: tactic field_output: state_after format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: datlaaaaaaa/e8cccbe1-c136-409a-a725-9f09127c1a3f hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/854bca96bed40197_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: cff9d1c5-a847-4707-b347-d0451baf6b24 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: cff9d1c5-a847-4707-b347-d0451baf6b24 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # e8cccbe1-c136-409a-a725-9f09127c1a3f This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.3004 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0849 | 0.0077 | 200 | 0.3004 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

adamo1139/DeepSeek-R1-Distill-Qwen-1.5B-6bpw-exl2

adamo1139

2025-01-20T23:21:56Z

8

0

null

[ "qwen2", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "6-bit", "exl2", "region:us" ]

null

2025-01-20T22:51:16Z

--- base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --- # DeepSeek-R1    <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- **Post-Training: Large-Scale Reinforcement Learning on the Base Model** - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- **Distillation: Smaller Models Can Be Powerful Too** - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> | **Model** | **#Total Params** | **#Activated Params** | **Context Length** | **Download** | | :------------: | :------------: | :------------: | :------------: | :------------: | | DeepSeek-R1-Zero | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) | | DeepSeek-R1 | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) | </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> | **Model** | **Base Model** | **Download** | | :------------: | :------------: | :------------: | | DeepSeek-R1-Distill-Qwen-1.5B | [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) | | DeepSeek-R1-Distill-Qwen-7B | [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) | | DeepSeek-R1-Distill-Llama-8B | [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) | | DeepSeek-R1-Distill-Qwen-14B | [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) | |DeepSeek-R1-Distill-Qwen-32B | [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) | | DeepSeek-R1-Distill-Llama-70B | [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) | </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> | Category | Benchmark (Metric) | Claude-3.5-Sonnet-1022 | GPT-4o 0513 | DeepSeek V3 | OpenAI o1-mini | OpenAI o1-1217 | DeepSeek R1 | |----------|-------------------|----------------------|------------|--------------|----------------|------------|--------------| | | Architecture | - | - | MoE | - | - | MoE | | | # Activated Params | - | - | 37B | - | - | 37B | | | # Total Params | - | - | 671B | - | - | 671B | | English | MMLU (Pass@1) | 88.3 | 87.2 | 88.5 | 85.2 | **91.8** | 90.8 | | | MMLU-Redux (EM) | 88.9 | 88.0 | 89.1 | 86.7 | - | **92.9** | | | MMLU-Pro (EM) | 78.0 | 72.6 | 75.9 | 80.3 | - | **84.0** | | | DROP (3-shot F1) | 88.3 | 83.7 | 91.6 | 83.9 | 90.2 | **92.2** | | | IF-Eval (Prompt Strict) | **86.5** | 84.3 | 86.1 | 84.8 | - | 83.3 | | | GPQA-Diamond (Pass@1) | 65.0 | 49.9 | 59.1 | 60.0 | **75.7** | 71.5 | | | SimpleQA (Correct) | 28.4 | 38.2 | 24.9 | 7.0 | **47.0** | 30.1 | | | FRAMES (Acc.) | 72.5 | 80.5 | 73.3 | 76.9 | - | **82.5** | | | AlpacaEval2.0 (LC-winrate) | 52.0 | 51.1 | 70.0 | 57.8 | - | **87.6** | | | ArenaHard (GPT-4-1106) | 85.2 | 80.4 | 85.5 | 92.0 | - | **92.3** | | Code | LiveCodeBench (Pass@1-COT) | 33.8 | 34.2 | - | 53.8 | 63.4 | **65.9** | | | Codeforces (Percentile) | 20.3 | 23.6 | 58.7 | 93.4 | **96.6** | 96.3 | | | Codeforces (Rating) | 717 | 759 | 1134 | 1820 | **2061** | 2029 | | | SWE Verified (Resolved) | **50.8** | 38.8 | 42.0 | 41.6 | 48.9 | 49.2 | | | Aider-Polyglot (Acc.) | 45.3 | 16.0 | 49.6 | 32.9 | **61.7** | 53.3 | | Math | AIME 2024 (Pass@1) | 16.0 | 9.3 | 39.2 | 63.6 | 79.2 | **79.8** | | | MATH-500 (Pass@1) | 78.3 | 74.6 | 90.2 | 90.0 | 96.4 | **97.3** | | | CNMO 2024 (Pass@1) | 13.1 | 10.8 | 43.2 | 67.6 | - | **78.8** | | Chinese | CLUEWSC (EM) | 85.4 | 87.9 | 90.9 | 89.9 | - | **92.8** | | | C-Eval (EM) | 76.7 | 76.0 | 86.5 | 68.9 | - | **91.8** | | | C-SimpleQA (Correct) | 55.4 | 58.7 | **68.0** | 40.3 | - | 63.7 | </div> ### Distilled Model Evaluation <div align="center"> | Model | AIME 2024 pass@1 | AIME 2024 cons@64 | MATH-500 pass@1 | GPQA Diamond pass@1 | LiveCodeBench pass@1 | CodeForces rating | |------------------------------------------|------------------|-------------------|-----------------|----------------------|----------------------|-------------------| | GPT-4o-0513 | 9.3 | 13.4 | 74.6 | 49.9 | 32.9 | 759 | | Claude-3.5-Sonnet-1022 | 16.0 | 26.7 | 78.3 | 65.0 | 38.9 | 717 | | o1-mini | 63.6 | 80.0 | 90.0 | 60.0 | 53.8 | **1820** | | QwQ-32B-Preview | 44.0 | 60.0 | 90.6 | 54.5 | 41.9 | 1316 | | DeepSeek-R1-Distill-Qwen-1.5B | 28.9 | 52.7 | 83.9 | 33.8 | 16.9 | 954 | | DeepSeek-R1-Distill-Qwen-7B | 55.5 | 83.3 | 92.8 | 49.1 | 37.6 | 1189 | | DeepSeek-R1-Distill-Qwen-14B | 69.7 | 80.0 | 93.9 | 59.1 | 53.1 | 1481 | | DeepSeek-R1-Distill-Qwen-32B | **72.6** | 83.3 | 94.3 | 62.1 | 57.2 | 1691 | | DeepSeek-R1-Distill-Llama-8B | 50.4 | 80.0 | 89.1 | 49.0 | 39.6 | 1205 | | DeepSeek-R1-Distill-Llama-70B | 70.0 | **86.7** | **94.5** | **65.2** | **57.5** | 1633 | </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` **NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output.** ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).

adamo1139/DeepSeek-R1-Distill-Qwen-1.5B-8bpw-exl2

adamo1139

2025-01-20T23:21:21Z

27

0

null

[ "qwen2", "base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "base_model:quantized:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", "8-bit", "exl2", "region:us" ]

null

2025-01-20T22:50:28Z

--- base_model: - deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --- # DeepSeek-R1    <div align="center"> <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek-V3" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="https://www.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Homepage" src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://chat.deepseek.com/" target="_blank" style="margin: 2px;"> <img alt="Chat" src="https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://huggingface.co/deepseek-ai" target="_blank" style="margin: 2px;"> <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://discord.gg/Tc7c45Zzu5" target="_blank" style="margin: 2px;"> <img alt="Discord" src="https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true" target="_blank" style="margin: 2px;"> <img alt="Wechat" src="https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://twitter.com/deepseek_ai" target="_blank" style="margin: 2px;"> <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div> <div align="center" style="line-height: 1;"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE" style="margin: 2px;"> <img alt="Code License" src="https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL" style="margin: 2px;"> <img alt="Model License" src="https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53" style="display: inline-block; vertical-align: middle;"/> </a> </div> <p align="center"> <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a> </p> ## 1. Introduction We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. <p align="center"> <img width="80%" src="figures/benchmark.jpg"> </p> ## 2. Model Summary --- **Post-Training: Large-Scale Reinforcement Learning on the Base Model** - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area. - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities. We believe the pipeline will benefit the industry by creating better models. --- **Distillation: Smaller Models Can Be Powerful Too** - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future. - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community. ## 3. Model Downloads ### DeepSeek-R1 Models <div align="center"> | **Model** | **#Total Params** | **#Activated Params** | **Context Length** | **Download** | | :------------: | :------------: | :------------: | :------------: | :------------: | | DeepSeek-R1-Zero | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) | | DeepSeek-R1 | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) | </div> DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base. For more details regrading the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository. ### DeepSeek-R1-Distill Models <div align="center"> | **Model** | **Base Model** | **Download** | | :------------: | :------------: | :------------: | | DeepSeek-R1-Distill-Qwen-1.5B | [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) | | DeepSeek-R1-Distill-Qwen-7B | [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) | | DeepSeek-R1-Distill-Llama-8B | [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) | | DeepSeek-R1-Distill-Qwen-14B | [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) | |DeepSeek-R1-Distill-Qwen-32B | [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) | | DeepSeek-R1-Distill-Llama-70B | [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) | </div> DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1. We slightly change their configs and tokenizers. Please use our setting to run these models. ## 4. Evaluation Results ### DeepSeek-R1-Evaluation For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1. <div align="center"> | Category | Benchmark (Metric) | Claude-3.5-Sonnet-1022 | GPT-4o 0513 | DeepSeek V3 | OpenAI o1-mini | OpenAI o1-1217 | DeepSeek R1 | |----------|-------------------|----------------------|------------|--------------|----------------|------------|--------------| | | Architecture | - | - | MoE | - | - | MoE | | | # Activated Params | - | - | 37B | - | - | 37B | | | # Total Params | - | - | 671B | - | - | 671B | | English | MMLU (Pass@1) | 88.3 | 87.2 | 88.5 | 85.2 | **91.8** | 90.8 | | | MMLU-Redux (EM) | 88.9 | 88.0 | 89.1 | 86.7 | - | **92.9** | | | MMLU-Pro (EM) | 78.0 | 72.6 | 75.9 | 80.3 | - | **84.0** | | | DROP (3-shot F1) | 88.3 | 83.7 | 91.6 | 83.9 | 90.2 | **92.2** | | | IF-Eval (Prompt Strict) | **86.5** | 84.3 | 86.1 | 84.8 | - | 83.3 | | | GPQA-Diamond (Pass@1) | 65.0 | 49.9 | 59.1 | 60.0 | **75.7** | 71.5 | | | SimpleQA (Correct) | 28.4 | 38.2 | 24.9 | 7.0 | **47.0** | 30.1 | | | FRAMES (Acc.) | 72.5 | 80.5 | 73.3 | 76.9 | - | **82.5** | | | AlpacaEval2.0 (LC-winrate) | 52.0 | 51.1 | 70.0 | 57.8 | - | **87.6** | | | ArenaHard (GPT-4-1106) | 85.2 | 80.4 | 85.5 | 92.0 | - | **92.3** | | Code | LiveCodeBench (Pass@1-COT) | 33.8 | 34.2 | - | 53.8 | 63.4 | **65.9** | | | Codeforces (Percentile) | 20.3 | 23.6 | 58.7 | 93.4 | **96.6** | 96.3 | | | Codeforces (Rating) | 717 | 759 | 1134 | 1820 | **2061** | 2029 | | | SWE Verified (Resolved) | **50.8** | 38.8 | 42.0 | 41.6 | 48.9 | 49.2 | | | Aider-Polyglot (Acc.) | 45.3 | 16.0 | 49.6 | 32.9 | **61.7** | 53.3 | | Math | AIME 2024 (Pass@1) | 16.0 | 9.3 | 39.2 | 63.6 | 79.2 | **79.8** | | | MATH-500 (Pass@1) | 78.3 | 74.6 | 90.2 | 90.0 | 96.4 | **97.3** | | | CNMO 2024 (Pass@1) | 13.1 | 10.8 | 43.2 | 67.6 | - | **78.8** | | Chinese | CLUEWSC (EM) | 85.4 | 87.9 | 90.9 | 89.9 | - | **92.8** | | | C-Eval (EM) | 76.7 | 76.0 | 86.5 | 68.9 | - | **91.8** | | | C-SimpleQA (Correct) | 55.4 | 58.7 | **68.0** | 40.3 | - | 63.7 | </div> ### Distilled Model Evaluation <div align="center"> | Model | AIME 2024 pass@1 | AIME 2024 cons@64 | MATH-500 pass@1 | GPQA Diamond pass@1 | LiveCodeBench pass@1 | CodeForces rating | |------------------------------------------|------------------|-------------------|-----------------|----------------------|----------------------|-------------------| | GPT-4o-0513 | 9.3 | 13.4 | 74.6 | 49.9 | 32.9 | 759 | | Claude-3.5-Sonnet-1022 | 16.0 | 26.7 | 78.3 | 65.0 | 38.9 | 717 | | o1-mini | 63.6 | 80.0 | 90.0 | 60.0 | 53.8 | **1820** | | QwQ-32B-Preview | 44.0 | 60.0 | 90.6 | 54.5 | 41.9 | 1316 | | DeepSeek-R1-Distill-Qwen-1.5B | 28.9 | 52.7 | 83.9 | 33.8 | 16.9 | 954 | | DeepSeek-R1-Distill-Qwen-7B | 55.5 | 83.3 | 92.8 | 49.1 | 37.6 | 1189 | | DeepSeek-R1-Distill-Qwen-14B | 69.7 | 80.0 | 93.9 | 59.1 | 53.1 | 1481 | | DeepSeek-R1-Distill-Qwen-32B | **72.6** | 83.3 | 94.3 | 62.1 | 57.2 | 1691 | | DeepSeek-R1-Distill-Llama-8B | 50.4 | 80.0 | 89.1 | 49.0 | 39.6 | 1205 | | DeepSeek-R1-Distill-Llama-70B | 70.0 | **86.7** | **94.5** | **65.2** | **57.5** | 1633 | </div> ## 5. Chat Website & API Platform You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button "DeepThink" We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/) ## 6. How to Run Locally ### DeepSeek-R1 Models Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally. ### DeepSeek-R1-Distill Models DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models. For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm): ```shell vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager ``` **NOTE: We recommend setting an appropriate temperature (between 0.5 and 0.7) when running these models, otherwise you may encounter issues with endless repetition or incoherent output.** ## 7. License This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE). DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that: - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1. - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE). - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE). ## 8. Citation ``` ``` ## 9. Contact If you have any questions, please raise an issue or contact us at [[email protected]]([email protected]).

LHRuig/cinestyle

LHRuig

2025-01-20T23:20:52Z

5

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:20:41Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # cinestyle <Gallery /> ## Model description cinestyle lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/cinestyle/tree/main) them in the Files & versions tab.

LHRuig/cinedrama

LHRuig

2025-01-20T23:20:46Z

8

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T23:19:20Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # cinedrama <Gallery /> ## Model description cinedrama lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/cinedrama/tree/main) them in the Files & versions tab.

kaizen9/phi-1_5_HQ_3000_20k

kaizen9

2025-01-20T23:20:32Z

15

0

transformers

[ "transformers", "safetensors", "phi", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]

text-generation

2025-01-20T23:16:36Z

--- library_name: transformers tags: [] --- # Model Card for Model ID  ## Model Details ### Model Description  This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional]  - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses  ### Direct Use  [More Information Needed] ### Downstream Use [optional]  [More Information Needed] ### Out-of-Scope Use  [More Information Needed] ## Bias, Risks, and Limitations  [More Information Needed] ### Recommendations  Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data  [More Information Needed] ### Training Procedure  #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed]  #### Speeds, Sizes, Times [optional]  [More Information Needed] ## Evaluation  ### Testing Data, Factors & Metrics #### Testing Data  [More Information Needed] #### Factors  [More Information Needed] #### Metrics  [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional]  [More Information Needed] ## Environmental Impact  Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional]  **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional]  [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]

lesso08/7bc9e54e-f755-40b7-a740-c391d742641d

lesso08

2025-01-20T23:20:20Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2-7B-Instruct", "base_model:adapter:unsloth/Qwen2-7B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:22:58Z

--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2-7B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 7bc9e54e-f755-40b7-a740-c391d742641d results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2-7B-Instruct bf16: true chat_template: llama3 datasets: - data_files: - 6e60a538f672529c_train_data.json ds_type: json format: custom path: /workspace/input_data/6e60a538f672529c_train_data.json type: field_input: communityName field_instruction: label field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso08/7bc9e54e-f755-40b7-a740-c391d742641d hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/6e60a538f672529c_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 3eb92360-4e77-4cc9-9ffa-0e03d7ea7423 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 3eb92360-4e77-4cc9-9ffa-0e03d7ea7423 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 7bc9e54e-f755-40b7-a740-c391d742641d This model is a fine-tuned version of [unsloth/Qwen2-7B-Instruct](https://huggingface.co/unsloth/Qwen2-7B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.0001 | 1 | nan | | 0.0 | 0.0004 | 5 | nan | | 0.0 | 0.0008 | 10 | nan | | 0.0 | 0.0012 | 15 | nan | | 0.0 | 0.0016 | 20 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

mrHungddddh/9b1deaac-4565-48f2-a511-89a1ab96e3e3

mrHungddddh

2025-01-20T23:19:50Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:04:02Z

--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 9b1deaac-4565-48f2-a511-89a1ab96e3e3 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - a7412d8c8f805ddf_train_data.json ds_type: json format: custom path: /workspace/input_data/a7412d8c8f805ddf_train_data.json type: field_instruction: premise field_output: hypothesis format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: mrHungddddh/9b1deaac-4565-48f2-a511-89a1ab96e3e3 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/a7412d8c8f805ddf_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ba28db81-f399-44e5-bdef-7af8dcf5a4ca warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 9b1deaac-4565-48f2-a511-89a1ab96e3e3 This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2632 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.7427 | 0.0304 | 200 | 1.2632 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

rawsh/q1-3B-PRIME

rawsh

2025-01-20T23:19:31Z

523

1

null

[ "safetensors", "qwen2", "text-generation", "conversational", "en", "dataset:PRIME-RL/Eurus-2-RL-Data", "base_model:PowerInfer/SmallThinker-3B-Preview", "base_model:finetune:PowerInfer/SmallThinker-3B-Preview", "region:us" ]

text-generation

2025-01-16T03:26:54Z

--- base_model: - Qwen/Qwen2.5-3B-Instruct - PowerInfer/SmallThinker-3B-Preview datasets: - PRIME-RL/Eurus-2-RL-Data language: - en pipeline_tag: text-generation --- # q1-3B-PRIME **q1-3B-PRIME**, a small reasoning model trained with reinforcement learning. Trained using SmallThinker-3B-Preview as a base model (Qwen2.5-3B-Instruct full finetuned on QwQ reasoning traces) for a roughly ~22.5% improvement on the test set in 120 training steps. (Note: lots of performance left on the table since PRIME saturates after 300+ steps.) # Benchmark Performance ## Math | Model | AIME24 | AMC23 | MATH-500 | |---------|--------|-------|-------| | Qwen2.5-3B-Instruct | 6.67 | 45 | - | | SmallThinker-3B-Preview| 16.667 | 57.5 | - | | **q1-3B-PRIME** | **26.667** | **67.5** | 64.8 | | Eurus-7B-PRIME | **26.667** | 57.8 | **79.2** | | GPT-4o | 9.3 | 45.8 | 76.4 | ## Coding | Model | HumanEval | Leetcode | |---------|--------|-------| | Qwen2.5-3B-Instruct | 74.4 | - | | **q1-3B-PRIME** | 71.95 | 20.55 | | GPT-4o | 90.2 | - |

laquythang/efe15045-778f-4dfa-8d03-17ab758f41b0

laquythang

2025-01-20T23:18:35Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:04:13Z

--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: efe15045-778f-4dfa-8d03-17ab758f41b0 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - a7412d8c8f805ddf_train_data.json ds_type: json format: custom path: /workspace/input_data/a7412d8c8f805ddf_train_data.json type: field_instruction: premise field_output: hypothesis format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: laquythang/efe15045-778f-4dfa-8d03-17ab758f41b0 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/a7412d8c8f805ddf_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ba28db81-f399-44e5-bdef-7af8dcf5a4ca warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # efe15045-778f-4dfa-8d03-17ab758f41b0 This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2649 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.7354 | 0.0304 | 200 | 1.2649 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

nadejdatarabukina/7623d9ae-b2cb-4a91-801e-e6ab04be6251

nadejdatarabukina

2025-01-20T23:18:27Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:unsloth/Meta-Llama-3.1-8B", "base_model:adapter:unsloth/Meta-Llama-3.1-8B", "license:llama3.1", "region:us" ]

null

2025-01-20T22:59:19Z

--- library_name: peft license: llama3.1 base_model: unsloth/Meta-Llama-3.1-8B tags: - axolotl - generated_from_trainer model-index: - name: 7623d9ae-b2cb-4a91-801e-e6ab04be6251 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Meta-Llama-3.1-8B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 91e193d3dca1611f_train_data.json ds_type: json format: custom path: /workspace/input_data/91e193d3dca1611f_train_data.json type: field_input: parent_id field_instruction: role field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: null eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: nadejdatarabukina/7623d9ae-b2cb-4a91-801e-e6ab04be6251 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 75GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/91e193d3dca1611f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 856a9aac-189f-40f7-b27c-c5616995b0d1 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 856a9aac-189f-40f7-b27c-c5616995b0d1 warmup_steps: 10 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 7623d9ae-b2cb-4a91-801e-e6ab04be6251 This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0004 | 1 | nan | | 0.0 | 0.0018 | 5 | nan | | 0.0 | 0.0035 | 10 | nan | | 0.0 | 0.0053 | 15 | nan | | 0.0 | 0.0071 | 20 | nan | | 0.0 | 0.0088 | 25 | nan | | 0.0 | 0.0106 | 30 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

John6666/real-dream-sdxlpony14-sdxl

John6666

2025-01-20T23:18:26Z

659

0

diffusers

[ "diffusers", "safetensors", "text-to-image", "stable-diffusion", "stable-diffusion-xl", "realistic", "photorealistic", "photo", "pony", "en", "base_model:luisrguerra/real-dream-xl-pony-releases", "base_model:finetune:luisrguerra/real-dream-xl-pony-releases", "license:other", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionXLPipeline", "region:us" ]

text-to-image

2025-01-20T23:13:33Z

--- license: other license_name: faipl-1.0-sd license_link: https://freedevproject.org/faipl-1.0-sd/ language: - en library_name: diffusers pipeline_tag: text-to-image tags: - text-to-image - stable-diffusion - stable-diffusion-xl - realistic - photorealistic - photo - pony base_model: luisrguerra/real-dream-xl-pony-releases --- Original model is [here](https://civitai.com/models/153568/real-dream?modelVersionId=1308507). The author is [here](https://huggingface.co/luisrguerra). This model created by [sinatra](https://civitai.com/user/sinatra).

nblinh/94055e20-ee90-4d03-9da3-41f4e7349a5a

nblinh

2025-01-20T23:17:56Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4", "base_model:adapter:MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T21:52:01Z

--- library_name: peft base_model: MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4 tags: - axolotl - generated_from_trainer model-index: - name: 94055e20-ee90-4d03-9da3-41f4e7349a5a results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 1983306ea4f53c9d_train_data.json ds_type: json format: custom path: /workspace/input_data/1983306ea4f53c9d_train_data.json type: field_input: prompt field_instruction: instruction field_output: response format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nblinh/94055e20-ee90-4d03-9da3-41f4e7349a5a hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/1983306ea4f53c9d_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 75177f7d-3059-4092-918d-8e9c49bae6b5 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 75177f7d-3059-4092-918d-8e9c49bae6b5 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 94055e20-ee90-4d03-9da3-41f4e7349a5a This model is a fine-tuned version of [MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4](https://huggingface.co/MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.3520 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.333 | 0.0334 | 200 | 0.3520 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

Monday-Someday/mobilenet_v2_1.0_224-finetuned-ISIC-dec2024test

Monday-Someday

2025-01-20T23:17:03Z

8

0

transformers

[ "transformers", "tensorboard", "safetensors", "mobilenet_v2", "image-classification", "generated_from_trainer", "dataset:imagefolder", "base_model:google/mobilenet_v2_1.0_224", "base_model:finetune:google/mobilenet_v2_1.0_224", "license:other", "model-index", "autotrain_compatible", "endpoints_compatible", "region:us" ]

image-classification

2025-01-20T02:12:01Z

--- library_name: transformers license: other base_model: google/mobilenet_v2_1.0_224 tags: - generated_from_trainer datasets: - imagefolder metrics: - accuracy model-index: - name: mobilenet_v2_1.0_224-finetuned-ISIC-dec2024test results: - task: name: Image Classification type: image-classification dataset: name: imagefolder type: imagefolder config: default split: train args: default metrics: - name: Accuracy type: accuracy value: 0.9276220745449292 ---  # mobilenet_v2_1.0_224-finetuned-ISIC-dec2024test This model is a fine-tuned version of [google/mobilenet_v2_1.0_224](https://huggingface.co/google/mobilenet_v2_1.0_224) on the imagefolder dataset. It achieves the following results on the evaluation set: - Loss: 0.1763 - Accuracy: 0.9276 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 32 - eval_batch_size: 32 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 128 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1 - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:------:|:----:|:---------------:|:--------:| | 0.9055 | 0.9985 | 486 | 0.1955 | 0.9195 | | 0.8797 | 1.9985 | 972 | 0.2074 | 0.9138 | | 0.8144 | 2.9985 | 1458 | 0.1797 | 0.9263 | | 0.9243 | 3.9985 | 1944 | 0.1862 | 0.9233 | | 0.8199 | 4.9985 | 2430 | 0.1763 | 0.9276 | ### Framework versions - Transformers 4.47.1 - Pytorch 2.5.1+cpu - Datasets 3.2.0 - Tokenizers 0.21.0

lesso15/858b3fe8-837d-4d0b-908d-af0eb85b1273

lesso15

2025-01-20T23:15:30Z

8

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/CodeLlama-13b-hf-flash", "base_model:adapter:NousResearch/CodeLlama-13b-hf-flash", "4-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:43:29Z

--- library_name: peft base_model: NousResearch/CodeLlama-13b-hf-flash tags: - axolotl - generated_from_trainer model-index: - name: 858b3fe8-837d-4d0b-908d-af0eb85b1273 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/CodeLlama-13b-hf-flash bf16: auto chat_template: llama3 datasets: - data_files: - 9ff4e3b24bf3b2a4_train_data.json ds_type: json format: custom path: /workspace/input_data/9ff4e3b24bf3b2a4_train_data.json type: field_input: sentence1 field_instruction: phrase1 field_output: sentence2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: true gradient_checkpointing: false group_by_length: false hub_model_id: lesso15/858b3fe8-837d-4d0b-908d-af0eb85b1273 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: true load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/9ff4e3b24bf3b2a4_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a warmup_steps: 10 weight_decay: 0.01 xformers_attention: null ``` </details><br> # 858b3fe8-837d-4d0b-908d-af0eb85b1273 This model is a fine-tuned version of [NousResearch/CodeLlama-13b-hf-flash](https://huggingface.co/NousResearch/CodeLlama-13b-hf-flash) on the None dataset. It achieves the following results on the evaluation set: - Loss: 3.0180 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0003 | 1 | 4.9637 | | 5.0445 | 0.0014 | 5 | 4.8259 | | 4.2696 | 0.0029 | 10 | 3.8969 | | 3.0239 | 0.0043 | 15 | 3.1104 | | 3.3092 | 0.0057 | 20 | 3.0548 | | 2.8685 | 0.0071 | 25 | 3.0180 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

kokovova/e5efb00d-850a-4f79-b6b3-19433c4b5d28

kokovova

2025-01-20T23:14:22Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:princeton-nlp/Sheared-LLaMA-1.3B", "base_model:adapter:princeton-nlp/Sheared-LLaMA-1.3B", "license:apache-2.0", "region:us" ]

null

2025-01-20T22:15:38Z

--- library_name: peft license: apache-2.0 base_model: princeton-nlp/Sheared-LLaMA-1.3B tags: - axolotl - generated_from_trainer model-index: - name: e5efb00d-850a-4f79-b6b3-19433c4b5d28 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: princeton-nlp/Sheared-LLaMA-1.3B bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 6d346ae45cb7310f_train_data.json ds_type: json format: custom path: /workspace/input_data/6d346ae45cb7310f_train_data.json type: field_input: problem field_instruction: prompt field_output: solution format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: kokovova/e5efb00d-850a-4f79-b6b3-19433c4b5d28 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/6d346ae45cb7310f_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 293f4171-e6c9-4854-a803-09018c88d137 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 293f4171-e6c9-4854-a803-09018c88d137 warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # e5efb00d-850a-4f79-b6b3-19433c4b5d28 This model is a fine-tuned version of [princeton-nlp/Sheared-LLaMA-1.3B](https://huggingface.co/princeton-nlp/Sheared-LLaMA-1.3B) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0000 | 1 | nan | | 0.0 | 0.0001 | 5 | nan | | 0.0 | 0.0002 | 10 | nan | | 0.0 | 0.0003 | 15 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

nhungphammmmm/d9757073-d7e1-40b9-be8d-0c20a4559179

nhungphammmmm

2025-01-20T23:13:54Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "base_model:adapter:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:02:39Z

--- library_name: peft base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f tags: - axolotl - generated_from_trainer model-index: - name: d9757073-d7e1-40b9-be8d-0c20a4559179 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json ds_type: json format: custom path: /workspace/input_data/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nhungphammmmm/d9757073-d7e1-40b9-be8d-0c20a4559179 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075526eb-32e0-4485-aab7-014e4d302171 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075526eb-32e0-4485-aab7-014e4d302171 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # d9757073-d7e1-40b9-be8d-0c20a4559179 This model is a fine-tuned version of [samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f](https://huggingface.co/samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.1318 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.2385 | 0.0407 | 200 | 1.1318 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

nhunglaaaaaaa/133de99e-2f53-4b9b-b8d2-edea6793573f

nhunglaaaaaaa

2025-01-20T23:13:15Z

8

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T23:03:55Z

--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 133de99e-2f53-4b9b-b8d2-edea6793573f results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen2.5-0.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - a7412d8c8f805ddf_train_data.json ds_type: json format: custom path: /workspace/input_data/a7412d8c8f805ddf_train_data.json type: field_instruction: premise field_output: hypothesis format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nhunglaaaaaaa/133de99e-2f53-4b9b-b8d2-edea6793573f hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/a7412d8c8f805ddf_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ba28db81-f399-44e5-bdef-7af8dcf5a4ca wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: ba28db81-f399-44e5-bdef-7af8dcf5a4ca warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 133de99e-2f53-4b9b-b8d2-edea6793573f This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2642 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.7557 | 0.0304 | 200 | 1.2642 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

sergioalves/94dd5542-37d0-4a0a-b042-711c0d084791

sergioalves

2025-01-20T23:12:29Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "base_model:adapter:samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f", "region:us" ]

null

2025-01-20T23:02:39Z

--- library_name: peft base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f tags: - axolotl - generated_from_trainer model-index: - name: 94dd5542-37d0-4a0a-b042-711c0d084791 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json ds_type: json format: custom path: /workspace/input_data/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json type: field_input: input field_instruction: instruction field_output: output format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: sergioalves/94dd5542-37d0-4a0a-b042-711c0d084791 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 75GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/train_c4393383-ef1d-4e9c-b95c-18b4f735570d.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_hf output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 075526eb-32e0-4485-aab7-014e4d302171 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 075526eb-32e0-4485-aab7-014e4d302171 warmup_steps: 10 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 94dd5542-37d0-4a0a-b042-711c0d084791 This model is a fine-tuned version of [samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f](https://huggingface.co/samoline/7d183bf9-ed95-443c-94dc-1cad850bf23f) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_HF with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0002 | 1 | nan | | 0.0 | 0.0010 | 5 | nan | | 0.0 | 0.0020 | 10 | nan | | 0.0 | 0.0031 | 15 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

LHRuig/hyperreal

LHRuig

2025-01-20T23:10:45Z

7

0

diffusers

[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]

text-to-image

2025-01-20T22:56:28Z

--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: suit output: url: images/suit.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: man --- # hyperreal <Gallery /> ## Model description hyperreal lora ## Trigger words You should use `man` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/LHRuig/hyperreal/tree/main) them in the Files & versions tab.

MayBashendy/ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k17_task5_organization

MayBashendy

2025-01-20T23:09:52Z

5

0

transformers

[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]

text-classification

2025-01-20T23:00:25Z

--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k17_task5_organization results: [] ---  # ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k17_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2639 - Qwk: 0.1814 - Mse: 1.2639 - Rmse: 1.1242 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | Qwk | Mse | Rmse | |:-------------:|:-------:|:----:|:---------------:|:-------:|:------:|:------:| | No log | 0.0488 | 2 | 3.9152 | -0.0319 | 3.9152 | 1.9787 | | No log | 0.0976 | 4 | 2.3161 | 0.0372 | 2.3161 | 1.5219 | | No log | 0.1463 | 6 | 2.0454 | 0.0260 | 2.0454 | 1.4302 | | No log | 0.1951 | 8 | 1.6301 | 0.0100 | 1.6301 | 1.2768 | | No log | 0.2439 | 10 | 1.2168 | 0.1091 | 1.2168 | 1.1031 | | No log | 0.2927 | 12 | 1.1692 | 0.1901 | 1.1692 | 1.0813 | | No log | 0.3415 | 14 | 1.2690 | 0.1205 | 1.2690 | 1.1265 | | No log | 0.3902 | 16 | 1.3094 | 0.0613 | 1.3094 | 1.1443 | | No log | 0.4390 | 18 | 1.2486 | 0.0374 | 1.2486 | 1.1174 | | No log | 0.4878 | 20 | 1.1393 | 0.1304 | 1.1393 | 1.0674 | | No log | 0.5366 | 22 | 1.0717 | 0.1167 | 1.0717 | 1.0352 | | No log | 0.5854 | 24 | 1.0785 | 0.0888 | 1.0785 | 1.0385 | | No log | 0.6341 | 26 | 1.1062 | 0.0445 | 1.1062 | 1.0518 | | No log | 0.6829 | 28 | 1.0674 | 0.1203 | 1.0674 | 1.0331 | | No log | 0.7317 | 30 | 0.9890 | 0.1107 | 0.9890 | 0.9945 | | No log | 0.7805 | 32 | 0.9718 | 0.3221 | 0.9718 | 0.9858 | | No log | 0.8293 | 34 | 1.1094 | 0.1576 | 1.1094 | 1.0533 | | No log | 0.8780 | 36 | 1.0090 | 0.2392 | 1.0090 | 1.0045 | | No log | 0.9268 | 38 | 0.9661 | 0.3935 | 0.9661 | 0.9829 | | No log | 0.9756 | 40 | 0.9632 | 0.2834 | 0.9632 | 0.9814 | | No log | 1.0244 | 42 | 1.0046 | 0.2217 | 1.0046 | 1.0023 | | No log | 1.0732 | 44 | 1.0990 | 0.1189 | 1.0990 | 1.0483 | | No log | 1.1220 | 46 | 1.1539 | 0.0761 | 1.1539 | 1.0742 | | No log | 1.1707 | 48 | 1.1020 | 0.0823 | 1.1020 | 1.0498 | | No log | 1.2195 | 50 | 1.0902 | 0.1350 | 1.0902 | 1.0441 | | No log | 1.2683 | 52 | 1.1294 | 0.1003 | 1.1294 | 1.0627 | | No log | 1.3171 | 54 | 1.1120 | 0.2145 | 1.1120 | 1.0545 | | No log | 1.3659 | 56 | 1.0491 | 0.2912 | 1.0491 | 1.0243 | | No log | 1.4146 | 58 | 0.9670 | 0.3048 | 0.9670 | 0.9834 | | No log | 1.4634 | 60 | 0.9772 | 0.1330 | 0.9772 | 0.9885 | | No log | 1.5122 | 62 | 1.0123 | 0.0888 | 1.0123 | 1.0061 | | No log | 1.5610 | 64 | 1.0275 | 0.0888 | 1.0275 | 1.0136 | | No log | 1.6098 | 66 | 1.0126 | 0.0888 | 1.0126 | 1.0063 | | No log | 1.6585 | 68 | 0.9315 | 0.2672 | 0.9315 | 0.9651 | | No log | 1.7073 | 70 | 0.9216 | 0.2895 | 0.9216 | 0.9600 | | No log | 1.7561 | 72 | 1.0374 | 0.2392 | 1.0374 | 1.0186 | | No log | 1.8049 | 74 | 1.0876 | 0.3283 | 1.0876 | 1.0429 | | No log | 1.8537 | 76 | 1.1482 | 0.3283 | 1.1482 | 1.0716 | | No log | 1.9024 | 78 | 1.2706 | 0.2926 | 1.2706 | 1.1272 | | No log | 1.9512 | 80 | 1.4393 | 0.2588 | 1.4393 | 1.1997 | | No log | 2.0 | 82 | 1.6033 | 0.2016 | 1.6033 | 1.2662 | | No log | 2.0488 | 84 | 1.7020 | 0.2026 | 1.7020 | 1.3046 | | No log | 2.0976 | 86 | 1.6749 | 0.1914 | 1.6749 | 1.2942 | | No log | 2.1463 | 88 | 1.5513 | 0.0973 | 1.5513 | 1.2455 | | No log | 2.1951 | 90 | 1.4070 | 0.2149 | 1.4070 | 1.1862 | | No log | 2.2439 | 92 | 1.4139 | 0.1362 | 1.4139 | 1.1891 | | No log | 2.2927 | 94 | 1.4569 | 0.1297 | 1.4569 | 1.2070 | | No log | 2.3415 | 96 | 1.3699 | 0.0976 | 1.3699 | 1.1704 | | No log | 2.3902 | 98 | 1.1940 | 0.1434 | 1.1940 | 1.0927 | | No log | 2.4390 | 100 | 1.0175 | 0.2263 | 1.0175 | 1.0087 | | No log | 2.4878 | 102 | 0.9743 | 0.1857 | 0.9743 | 0.9871 | | No log | 2.5366 | 104 | 0.9971 | 0.1707 | 0.9971 | 0.9986 | | No log | 2.5854 | 106 | 1.0930 | 0.1970 | 1.0930 | 1.0454 | | No log | 2.6341 | 108 | 1.1941 | 0.0571 | 1.1941 | 1.0928 | | No log | 2.6829 | 110 | 1.3024 | -0.0297 | 1.3024 | 1.1412 | | No log | 2.7317 | 112 | 1.3887 | -0.0460 | 1.3887 | 1.1784 | | No log | 2.7805 | 114 | 1.4502 | 0.1194 | 1.4502 | 1.2043 | | No log | 2.8293 | 116 | 1.4877 | 0.2313 | 1.4877 | 1.2197 | | No log | 2.8780 | 118 | 1.5875 | 0.2638 | 1.5875 | 1.2599 | | No log | 2.9268 | 120 | 1.6965 | 0.2252 | 1.6965 | 1.3025 | | No log | 2.9756 | 122 | 1.7232 | 0.1078 | 1.7232 | 1.3127 | | No log | 3.0244 | 124 | 1.6639 | 0.1207 | 1.6639 | 1.2899 | | No log | 3.0732 | 126 | 1.5461 | 0.2004 | 1.5461 | 1.2434 | | No log | 3.1220 | 128 | 1.4334 | 0.1438 | 1.4334 | 1.1973 | | No log | 3.1707 | 130 | 1.4259 | 0.0712 | 1.4259 | 1.1941 | | No log | 3.2195 | 132 | 1.4356 | -0.0541 | 1.4356 | 1.1982 | | No log | 3.2683 | 134 | 1.4579 | -0.0939 | 1.4579 | 1.2074 | | No log | 3.3171 | 136 | 1.4491 | 0.0147 | 1.4491 | 1.2038 | | No log | 3.3659 | 138 | 1.5209 | 0.2752 | 1.5209 | 1.2332 | | No log | 3.4146 | 140 | 1.5553 | 0.2110 | 1.5553 | 1.2471 | | No log | 3.4634 | 142 | 1.5332 | 0.2317 | 1.5332 | 1.2382 | | No log | 3.5122 | 144 | 1.4563 | 0.2437 | 1.4563 | 1.2068 | | No log | 3.5610 | 146 | 1.4090 | 0.2694 | 1.4090 | 1.1870 | | No log | 3.6098 | 148 | 1.4642 | 0.2694 | 1.4642 | 1.2101 | | No log | 3.6585 | 150 | 1.6547 | 0.2644 | 1.6547 | 1.2864 | | No log | 3.7073 | 152 | 1.9515 | 0.2247 | 1.9515 | 1.3970 | | No log | 3.7561 | 154 | 2.0626 | 0.1896 | 2.0626 | 1.4362 | | No log | 3.8049 | 156 | 1.9816 | 0.2127 | 1.9816 | 1.4077 | | No log | 3.8537 | 158 | 1.7125 | 0.2252 | 1.7125 | 1.3086 | | No log | 3.9024 | 160 | 1.4900 | 0.2221 | 1.4900 | 1.2206 | | No log | 3.9512 | 162 | 1.3005 | 0.2424 | 1.3005 | 1.1404 | | No log | 4.0 | 164 | 1.3190 | 0.2709 | 1.3190 | 1.1485 | | No log | 4.0488 | 166 | 1.3899 | 0.2752 | 1.3899 | 1.1789 | | No log | 4.0976 | 168 | 1.5366 | 0.2363 | 1.5366 | 1.2396 | | No log | 4.1463 | 170 | 1.6811 | 0.2448 | 1.6811 | 1.2966 | | No log | 4.1951 | 172 | 1.6823 | 0.2406 | 1.6823 | 1.2971 | | No log | 4.2439 | 174 | 1.6922 | 0.2406 | 1.6922 | 1.3009 | | No log | 4.2927 | 176 | 1.6000 | 0.2869 | 1.6000 | 1.2649 | | No log | 4.3415 | 178 | 1.5276 | 0.2869 | 1.5276 | 1.2360 | | No log | 4.3902 | 180 | 1.5637 | 0.2869 | 1.5637 | 1.2505 | | No log | 4.4390 | 182 | 1.5011 | 0.2869 | 1.5011 | 1.2252 | | No log | 4.4878 | 184 | 1.3543 | 0.2982 | 1.3543 | 1.1637 | | No log | 4.5366 | 186 | 1.1786 | 0.2709 | 1.1786 | 1.0856 | | No log | 4.5854 | 188 | 1.1007 | 0.2795 | 1.1007 | 1.0492 | | No log | 4.6341 | 190 | 1.1784 | 0.2709 | 1.1784 | 1.0855 | | No log | 4.6829 | 192 | 1.3811 | 0.2982 | 1.3811 | 1.1752 | | No log | 4.7317 | 194 | 1.5727 | 0.2832 | 1.5727 | 1.2541 | | No log | 4.7805 | 196 | 1.6856 | 0.2270 | 1.6856 | 1.2983 | | No log | 4.8293 | 198 | 1.6794 | 0.2437 | 1.6794 | 1.2959 | | No log | 4.8780 | 200 | 1.5514 | 0.2522 | 1.5514 | 1.2456 | | No log | 4.9268 | 202 | 1.4239 | 0.2709 | 1.4239 | 1.1933 | | No log | 4.9756 | 204 | 1.2668 | 0.1142 | 1.2668 | 1.1255 | | No log | 5.0244 | 206 | 1.2415 | 0.1142 | 1.2415 | 1.1142 | | No log | 5.0732 | 208 | 1.3106 | 0.2065 | 1.3106 | 1.1448 | | No log | 5.1220 | 210 | 1.4415 | 0.2424 | 1.4415 | 1.2006 | | No log | 5.1707 | 212 | 1.6183 | 0.1832 | 1.6183 | 1.2721 | | No log | 5.2195 | 214 | 1.7023 | 0.1533 | 1.7023 | 1.3047 | | No log | 5.2683 | 216 | 1.6420 | 0.1142 | 1.6420 | 1.2814 | | No log | 5.3171 | 218 | 1.5048 | 0.1562 | 1.5048 | 1.2267 | | No log | 5.3659 | 220 | 1.4030 | 0.1744 | 1.4030 | 1.1845 | | No log | 5.4146 | 222 | 1.3956 | 0.1744 | 1.3956 | 1.1814 | | No log | 5.4634 | 224 | 1.4510 | 0.2065 | 1.4510 | 1.2046 | | No log | 5.5122 | 226 | 1.5505 | 0.2522 | 1.5505 | 1.2452 | | No log | 5.5610 | 228 | 1.5876 | 0.2568 | 1.5876 | 1.2600 | | No log | 5.6098 | 230 | 1.4723 | 0.1880 | 1.4723 | 1.2134 | | No log | 5.6585 | 232 | 1.4053 | 0.1562 | 1.4053 | 1.1854 | | No log | 5.7073 | 234 | 1.4550 | 0.1943 | 1.4550 | 1.2063 | | No log | 5.7561 | 236 | 1.5212 | 0.1943 | 1.5212 | 1.2334 | | No log | 5.8049 | 238 | 1.5631 | 0.1634 | 1.5631 | 1.2502 | | No log | 5.8537 | 240 | 1.6752 | 0.1141 | 1.6752 | 1.2943 | | No log | 5.9024 | 242 | 1.6585 | 0.1634 | 1.6585 | 1.2878 | | No log | 5.9512 | 244 | 1.5822 | 0.2004 | 1.5822 | 1.2578 | | No log | 6.0 | 246 | 1.4101 | 0.2522 | 1.4101 | 1.1875 | | No log | 6.0488 | 248 | 1.2845 | 0.2184 | 1.2845 | 1.1333 | | No log | 6.0976 | 250 | 1.3726 | 0.3052 | 1.3726 | 1.1716 | | No log | 6.1463 | 252 | 1.5895 | 0.2527 | 1.5895 | 1.2608 | | No log | 6.1951 | 254 | 1.7285 | 0.2419 | 1.7285 | 1.3147 | | No log | 6.2439 | 256 | 1.7297 | 0.2419 | 1.7297 | 1.3152 | | No log | 6.2927 | 258 | 1.5794 | 0.2296 | 1.5794 | 1.2567 | | No log | 6.3415 | 260 | 1.5724 | 0.2296 | 1.5724 | 1.2539 | | No log | 6.3902 | 262 | 1.5028 | 0.3052 | 1.5028 | 1.2259 | | No log | 6.4390 | 264 | 1.4208 | 0.2474 | 1.4208 | 1.1920 | | No log | 6.4878 | 266 | 1.3507 | 0.2126 | 1.3507 | 1.1622 | | No log | 6.5366 | 268 | 1.3940 | 0.2424 | 1.3940 | 1.1807 | | No log | 6.5854 | 270 | 1.5438 | 0.2611 | 1.5438 | 1.2425 | | No log | 6.6341 | 272 | 1.6434 | 0.2110 | 1.6434 | 1.2819 | | No log | 6.6829 | 274 | 1.6874 | 0.1752 | 1.6874 | 1.2990 | | No log | 6.7317 | 276 | 1.5767 | 0.2117 | 1.5767 | 1.2557 | | No log | 6.7805 | 278 | 1.4146 | 0.1486 | 1.4146 | 1.1893 | | No log | 6.8293 | 280 | 1.2838 | 0.1142 | 1.2838 | 1.1330 | | No log | 6.8780 | 282 | 1.2129 | 0.0781 | 1.2129 | 1.1013 | | No log | 6.9268 | 284 | 1.2424 | 0.1142 | 1.2424 | 1.1146 | | No log | 6.9756 | 286 | 1.3719 | 0.2424 | 1.3719 | 1.1713 | | No log | 7.0244 | 288 | 1.5464 | 0.2437 | 1.5464 | 1.2435 | | No log | 7.0732 | 290 | 1.7671 | 0.2007 | 1.7671 | 1.3293 | | No log | 7.1220 | 292 | 1.8037 | 0.1688 | 1.8037 | 1.3430 | | No log | 7.1707 | 294 | 1.7056 | 0.2058 | 1.7056 | 1.3060 | | No log | 7.2195 | 296 | 1.5468 | 0.2239 | 1.5468 | 1.2437 | | No log | 7.2683 | 298 | 1.3254 | 0.2126 | 1.3254 | 1.1513 | | No log | 7.3171 | 300 | 1.2248 | 0.1142 | 1.2248 | 1.1067 | | No log | 7.3659 | 302 | 1.2181 | 0.1024 | 1.2181 | 1.1037 | | No log | 7.4146 | 304 | 1.3199 | 0.2752 | 1.3199 | 1.1489 | | No log | 7.4634 | 306 | 1.4566 | 0.2869 | 1.4566 | 1.2069 | | No log | 7.5122 | 308 | 1.5103 | 0.2906 | 1.5103 | 1.2289 | | No log | 7.5610 | 310 | 1.5202 | 0.2733 | 1.5202 | 1.2330 | | No log | 7.6098 | 312 | 1.4770 | 0.3177 | 1.4770 | 1.2153 | | No log | 7.6585 | 314 | 1.4998 | 0.3205 | 1.4998 | 1.2247 | | No log | 7.7073 | 316 | 1.5688 | 0.2874 | 1.5688 | 1.2525 | | No log | 7.7561 | 318 | 1.6757 | 0.2681 | 1.6757 | 1.2945 | | No log | 7.8049 | 320 | 1.6044 | 0.2482 | 1.6044 | 1.2667 | | No log | 7.8537 | 322 | 1.4328 | 0.2522 | 1.4328 | 1.1970 | | No log | 7.9024 | 324 | 1.2480 | 0.1744 | 1.2480 | 1.1171 | | No log | 7.9512 | 326 | 1.1255 | 0.0401 | 1.1255 | 1.0609 | | No log | 8.0 | 328 | 1.0764 | 0.0781 | 1.0764 | 1.0375 | | No log | 8.0488 | 330 | 1.1043 | 0.1744 | 1.1043 | 1.0508 | | No log | 8.0976 | 332 | 1.2622 | 0.3018 | 1.2622 | 1.1235 | | No log | 8.1463 | 334 | 1.5303 | 0.2606 | 1.5303 | 1.2370 | | No log | 8.1951 | 336 | 1.7740 | 0.3024 | 1.7740 | 1.3319 | | No log | 8.2439 | 338 | 1.8534 | 0.2988 | 1.8534 | 1.3614 | | No log | 8.2927 | 340 | 1.7466 | 0.2967 | 1.7466 | 1.3216 | | No log | 8.3415 | 342 | 1.5761 | 0.2940 | 1.5761 | 1.2554 | | No log | 8.3902 | 344 | 1.4872 | 0.2793 | 1.4872 | 1.2195 | | No log | 8.4390 | 346 | 1.4299 | 0.2665 | 1.4299 | 1.1958 | | No log | 8.4878 | 348 | 1.4240 | 0.2665 | 1.4240 | 1.1933 | | No log | 8.5366 | 350 | 1.4802 | 0.2474 | 1.4802 | 1.2166 | | No log | 8.5854 | 352 | 1.4316 | 0.2709 | 1.4316 | 1.1965 | | No log | 8.6341 | 354 | 1.3518 | 0.2665 | 1.3518 | 1.1627 | | No log | 8.6829 | 356 | 1.4224 | 0.2568 | 1.4224 | 1.1926 | | No log | 8.7317 | 358 | 1.4891 | 0.3117 | 1.4891 | 1.2203 | | No log | 8.7805 | 360 | 1.4446 | 0.2709 | 1.4446 | 1.2019 | | No log | 8.8293 | 362 | 1.3826 | 0.2665 | 1.3826 | 1.1759 | | No log | 8.8780 | 364 | 1.3559 | 0.2665 | 1.3559 | 1.1644 | | No log | 8.9268 | 366 | 1.3007 | 0.1814 | 1.3007 | 1.1405 | | No log | 8.9756 | 368 | 1.3466 | 0.2372 | 1.3466 | 1.1604 | | No log | 9.0244 | 370 | 1.4908 | 0.2793 | 1.4908 | 1.2210 | | No log | 9.0732 | 372 | 1.5425 | 0.3052 | 1.5425 | 1.2420 | | No log | 9.1220 | 374 | 1.4719 | 0.2752 | 1.4719 | 1.2132 | | No log | 9.1707 | 376 | 1.3776 | 0.2126 | 1.3776 | 1.1737 | | No log | 9.2195 | 378 | 1.3444 | 0.1814 | 1.3444 | 1.1595 | | No log | 9.2683 | 380 | 1.3261 | 0.1814 | 1.3261 | 1.1516 | | No log | 9.3171 | 382 | 1.3477 | 0.2126 | 1.3477 | 1.1609 | | No log | 9.3659 | 384 | 1.3544 | 0.2126 | 1.3544 | 1.1638 | | No log | 9.4146 | 386 | 1.3364 | 0.2126 | 1.3364 | 1.1560 | | No log | 9.4634 | 388 | 1.3280 | 0.2126 | 1.3280 | 1.1524 | | No log | 9.5122 | 390 | 1.3023 | 0.2424 | 1.3023 | 1.1412 | | No log | 9.5610 | 392 | 1.3131 | 0.2944 | 1.3131 | 1.1459 | | No log | 9.6098 | 394 | 1.3688 | 0.2982 | 1.3688 | 1.1700 | | No log | 9.6585 | 396 | 1.4065 | 0.3018 | 1.4065 | 1.1860 | | No log | 9.7073 | 398 | 1.4039 | 0.3018 | 1.4039 | 1.1849 | | No log | 9.7561 | 400 | 1.3519 | 0.2982 | 1.3519 | 1.1627 | | No log | 9.8049 | 402 | 1.3142 | 0.2665 | 1.3142 | 1.1464 | | No log | 9.8537 | 404 | 1.3158 | 0.2665 | 1.3158 | 1.1471 | | No log | 9.9024 | 406 | 1.3620 | 0.2665 | 1.3620 | 1.1671 | | No log | 9.9512 | 408 | 1.4741 | 0.2832 | 1.4741 | 1.2141 | | No log | 10.0 | 410 | 1.4998 | 0.2832 | 1.4998 | 1.2247 | | No log | 10.0488 | 412 | 1.5107 | 0.2832 | 1.5107 | 1.2291 | | No log | 10.0976 | 414 | 1.5279 | 0.2832 | 1.5279 | 1.2361 | | No log | 10.1463 | 416 | 1.5454 | 0.2793 | 1.5454 | 1.2431 | | No log | 10.1951 | 418 | 1.5071 | 0.2793 | 1.5071 | 1.2276 | | No log | 10.2439 | 420 | 1.4512 | 0.2982 | 1.4512 | 1.2047 | | No log | 10.2927 | 422 | 1.3641 | 0.2982 | 1.3641 | 1.1680 | | No log | 10.3415 | 424 | 1.2933 | 0.2709 | 1.2933 | 1.1372 | | No log | 10.3902 | 426 | 1.2873 | 0.2424 | 1.2873 | 1.1346 | | No log | 10.4390 | 428 | 1.3117 | 0.2424 | 1.3117 | 1.1453 | | No log | 10.4878 | 430 | 1.3510 | 0.2424 | 1.3510 | 1.1623 | | No log | 10.5366 | 432 | 1.4179 | 0.2239 | 1.4179 | 1.1907 | | No log | 10.5854 | 434 | 1.4578 | 0.2832 | 1.4578 | 1.2074 | | No log | 10.6341 | 436 | 1.4536 | 0.2568 | 1.4536 | 1.2057 | | No log | 10.6829 | 438 | 1.4186 | 0.2424 | 1.4186 | 1.1911 | | No log | 10.7317 | 440 | 1.4318 | 0.2126 | 1.4318 | 1.1966 | | No log | 10.7805 | 442 | 1.4254 | 0.2126 | 1.4254 | 1.1939 | | No log | 10.8293 | 444 | 1.4542 | 0.2126 | 1.4542 | 1.2059 | | No log | 10.8780 | 446 | 1.4625 | 0.2239 | 1.4625 | 1.2094 | | No log | 10.9268 | 448 | 1.5026 | 0.1880 | 1.5026 | 1.2258 | | No log | 10.9756 | 450 | 1.4913 | 0.1562 | 1.4913 | 1.2212 | | No log | 11.0244 | 452 | 1.5060 | 0.1024 | 1.5060 | 1.2272 | | No log | 11.0732 | 454 | 1.5108 | 0.1744 | 1.5108 | 1.2291 | | No log | 11.1220 | 456 | 1.5158 | 0.2004 | 1.5158 | 1.2312 | | No log | 11.1707 | 458 | 1.5287 | 0.2522 | 1.5287 | 1.2364 | | No log | 11.2195 | 460 | 1.4799 | 0.2832 | 1.4799 | 1.2165 | | No log | 11.2683 | 462 | 1.3876 | 0.2982 | 1.3876 | 1.1780 | | No log | 11.3171 | 464 | 1.2905 | 0.3072 | 1.2905 | 1.1360 | | No log | 11.3659 | 466 | 1.2601 | 0.3072 | 1.2601 | 1.1226 | | No log | 11.4146 | 468 | 1.3082 | 0.2709 | 1.3082 | 1.1438 | | No log | 11.4634 | 470 | 1.3823 | 0.2982 | 1.3823 | 1.1757 | | No log | 11.5122 | 472 | 1.4097 | 0.2832 | 1.4097 | 1.1873 | | No log | 11.5610 | 474 | 1.3280 | 0.2982 | 1.3280 | 1.1524 | | No log | 11.6098 | 476 | 1.2140 | 0.3460 | 1.2140 | 1.1018 | | No log | 11.6585 | 478 | 1.1924 | 0.2730 | 1.1924 | 1.0920 | | No log | 11.7073 | 480 | 1.2291 | 0.2837 | 1.2291 | 1.1086 | | No log | 11.7561 | 482 | 1.3147 | 0.2869 | 1.3147 | 1.1466 | | No log | 11.8049 | 484 | 1.3998 | 0.2694 | 1.3998 | 1.1831 | | No log | 11.8537 | 486 | 1.4945 | 0.3429 | 1.4945 | 1.2225 | | No log | 11.9024 | 488 | 1.4541 | 0.3429 | 1.4540 | 1.2058 | | No log | 11.9512 | 490 | 1.3182 | 0.3086 | 1.3182 | 1.1481 | | No log | 12.0 | 492 | 1.1752 | 0.2877 | 1.1752 | 1.0841 | | No log | 12.0488 | 494 | 1.1071 | 0.3355 | 1.1071 | 1.0522 | | No log | 12.0976 | 496 | 1.1069 | 0.3231 | 1.1069 | 1.0521 | | No log | 12.1463 | 498 | 1.1869 | 0.3052 | 1.1869 | 1.0895 | | 0.3045 | 12.1951 | 500 | 1.3081 | 0.3329 | 1.3081 | 1.1437 | | 0.3045 | 12.2439 | 502 | 1.4234 | 0.3429 | 1.4234 | 1.1931 | | 0.3045 | 12.2927 | 504 | 1.4174 | 0.3329 | 1.4174 | 1.1906 | | 0.3045 | 12.3415 | 506 | 1.3559 | 0.3086 | 1.3559 | 1.1644 | | 0.3045 | 12.3902 | 508 | 1.2393 | 0.2126 | 1.2393 | 1.1132 | | 0.3045 | 12.4390 | 510 | 1.2199 | 0.1814 | 1.2199 | 1.1045 | | 0.3045 | 12.4878 | 512 | 1.2639 | 0.1814 | 1.2639 | 1.1242 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1

VERSIL91/13889a03-e443-4bcb-a2ab-46cfe5ea650a

VERSIL91

2025-01-20T23:08:56Z

6

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2-1.5B-Instruct", "base_model:adapter:unsloth/Qwen2-1.5B-Instruct", "license:apache-2.0", "region:us" ]

null

2025-01-20T23:08:51Z

--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2-1.5B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml accelerate_config: dynamo_backend: inductor mixed_precision: bf16 num_machines: 1 num_processes: auto use_cpu: false adapter: lora base_model: unsloth/Qwen2-1.5B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - dd06633aceb12410_train_data.json ds_type: json format: custom path: /workspace/input_data/dd06633aceb12410_train_data.json type: field_instruction: tests field_output: prompt format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 16 gradient_checkpointing: true group_by_length: false hub_model_id: null hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lora_target_modules: - q_proj - v_proj lr_scheduler: cosine max_memory: 0: 70GiB max_steps: 50 micro_batch_size: 2 mlflow_experiment_name: /tmp/dd06633aceb12410_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true quantization_config: llm_int8_enable_fp32_cpu_offload: true load_in_8bit: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 strict: false tf32: false tokenizer_type: AutoTokenizer torch_compile: true train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 370ef635-02c6-4a8f-be9e-f46f2205d9d9 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 370ef635-02c6-4a8f-be9e-f46f2205d9d9 This model is a fine-tuned version of [unsloth/Qwen2-1.5B-Instruct](https://huggingface.co/unsloth/Qwen2-1.5B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 16 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 2 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.7619 | 1 | nan | | 0.0 | 1.5238 | 2 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

John6666/pancake-mix-illustrious-sdxl

John6666

2025-01-20T23:08:37Z

67

0

diffusers

[ "diffusers", "safetensors", "text-to-image", "stable-diffusion", "stable-diffusion-xl", "anime", "girls", "characters", "illustrious", "en", "base_model:OnomaAIResearch/Illustrious-xl-early-release-v0", "base_model:finetune:OnomaAIResearch/Illustrious-xl-early-release-v0", "license:other", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionXLPipeline", "region:us" ]

text-to-image

2025-01-20T23:02:05Z

--- license: other license_name: faipl-1.0-sd license_link: https://freedevproject.org/faipl-1.0-sd/ language: - en library_name: diffusers pipeline_tag: text-to-image tags: - text-to-image - stable-diffusion - stable-diffusion-xl - anime - girls - characters - illustrious base_model: OnomaAIResearch/Illustrious-xl-early-release-v0 --- Original model is [here](https://civitai.com/models/896658?modelVersionId=1308210). This model created by [Sukizou](https://civitai.com/user/Sukizou).

andrewmalk/kitsman

andrewmalk

2025-01-20T23:08:11Z

65

0

diffusers

[ "diffusers", "flux", "lora", "replicate", "text-to-image", "en", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "license:other", "region:us" ]

text-to-image

2025-01-17T05:27:16Z

--- license: other license_name: flux-1-dev-non-commercial-license license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md language: - en tags: - flux - diffusers - lora - replicate base_model: "black-forest-labs/FLUX.1-dev" pipeline_tag: text-to-image # widget: # - text: >- # prompt # output: # url: https://... instance_prompt: okitsman --- # Kitsman <Gallery /> Trained on Replicate using: https://replicate.com/ostris/flux-dev-lora-trainer/train ## Trigger words You should use `okitsman` to trigger the image generation. ## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers) ```py from diffusers import AutoPipelineForText2Image import torch pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda') pipeline.load_lora_weights('andrewmalk/kitsman', weight_name='lora.safetensors') image = pipeline('your prompt').images[0] ``` For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)

lesso02/1a5098f7-cebd-4eeb-a0f1-0753bde57ace

lesso02

2025-01-20T23:07:05Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/CodeLlama-13b-hf-flash", "base_model:adapter:NousResearch/CodeLlama-13b-hf-flash", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:52:47Z

--- library_name: peft base_model: NousResearch/CodeLlama-13b-hf-flash tags: - axolotl - generated_from_trainer model-index: - name: 1a5098f7-cebd-4eeb-a0f1-0753bde57ace results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/CodeLlama-13b-hf-flash bf16: true chat_template: llama3 datasets: - data_files: - 9ff4e3b24bf3b2a4_train_data.json ds_type: json format: custom path: /workspace/input_data/9ff4e3b24bf3b2a4_train_data.json type: field_input: sentence1 field_instruction: phrase1 field_output: sentence2 format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: 2 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null flash_attention: false fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: false group_by_length: false hub_model_id: lesso02/1a5098f7-cebd-4eeb-a0f1-0753bde57ace hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 25 micro_batch_size: 2 mlflow_experiment_name: /tmp/9ff4e3b24bf3b2a4_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 05245b1d-e8ff-44bb-a139-f31fd23d5a4a warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # 1a5098f7-cebd-4eeb-a0f1-0753bde57ace This model is a fine-tuned version of [NousResearch/CodeLlama-13b-hf-flash](https://huggingface.co/NousResearch/CodeLlama-13b-hf-flash) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.4527 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 25 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 16.5546 | 0.0011 | 1 | 4.0087 | | 14.6587 | 0.0057 | 5 | 3.9323 | | 11.4912 | 0.0114 | 10 | 2.9548 | | 11.1484 | 0.0171 | 15 | 2.5602 | | 9.782 | 0.0229 | 20 | 2.4786 | | 8.6338 | 0.0286 | 25 | 2.4527 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

mrHunghddddd/71740281-efde-40d2-bc98-1a144c2a49c5

mrHunghddddd

2025-01-20T23:06:46Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:Vikhrmodels/Vikhr-7B-instruct_0.4", "base_model:adapter:Vikhrmodels/Vikhr-7B-instruct_0.4", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:22:29Z

--- library_name: peft base_model: Vikhrmodels/Vikhr-7B-instruct_0.4 tags: - axolotl - generated_from_trainer model-index: - name: 71740281-efde-40d2-bc98-1a144c2a49c5 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: Vikhrmodels/Vikhr-7B-instruct_0.4 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - cae8a8291b672052_train_data.json ds_type: json format: custom path: /workspace/input_data/cae8a8291b672052_train_data.json type: field_input: poem_meter field_instruction: poem_title field_output: poem_verses format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: mrHunghddddd/71740281-efde-40d2-bc98-1a144c2a49c5 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/cae8a8291b672052_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: d2d62167-223e-438b-b1e7-02a477624b1a wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: d2d62167-223e-438b-b1e7-02a477624b1a warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 71740281-efde-40d2-bc98-1a144c2a49c5 This model is a fine-tuned version of [Vikhrmodels/Vikhr-7B-instruct_0.4](https://huggingface.co/Vikhrmodels/Vikhr-7B-instruct_0.4) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.7143 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.7516 | 0.1941 | 200 | 1.7143 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

MayBashendy/ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k14_task5_organization

MayBashendy

2025-01-20T23:05:46Z

7

0

transformers

[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:aubmindlab/bert-base-arabertv02", "base_model:finetune:aubmindlab/bert-base-arabertv02", "autotrain_compatible", "endpoints_compatible", "region:us" ]

text-classification

2025-01-20T15:09:27Z

--- library_name: transformers base_model: aubmindlab/bert-base-arabertv02 tags: - generated_from_trainer model-index: - name: ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k14_task5_organization results: [] ---  # ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run1_AugV5_k14_task5_organization This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.6862 - Qwk: 0.5472 - Mse: 0.6862 - Rmse: 0.8284 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | Qwk | Mse | Rmse | |:-------------:|:-------:|:----:|:---------------:|:-------:|:------:|:------:| | No log | 0.0435 | 2 | 3.8865 | -0.0294 | 3.8865 | 1.9714 | | No log | 0.0870 | 4 | 2.0120 | 0.0142 | 2.0120 | 1.4184 | | No log | 0.1304 | 6 | 1.7353 | -0.0458 | 1.7353 | 1.3173 | | No log | 0.1739 | 8 | 1.2631 | 0.0380 | 1.2631 | 1.1239 | | No log | 0.2174 | 10 | 1.1664 | -0.0032 | 1.1664 | 1.0800 | | No log | 0.2609 | 12 | 1.1818 | 0.0380 | 1.1818 | 1.0871 | | No log | 0.3043 | 14 | 1.2235 | 0.0380 | 1.2235 | 1.1061 | | No log | 0.3478 | 16 | 1.2637 | 0.0380 | 1.2637 | 1.1242 | | No log | 0.3913 | 18 | 1.2607 | 0.0760 | 1.2607 | 1.1228 | | No log | 0.4348 | 20 | 1.2646 | 0.1142 | 1.2646 | 1.1246 | | No log | 0.4783 | 22 | 1.1968 | 0.1910 | 1.1968 | 1.0940 | | No log | 0.5217 | 24 | 1.1106 | 0.1805 | 1.1106 | 1.0539 | | No log | 0.5652 | 26 | 1.1018 | 0.1493 | 1.1018 | 1.0497 | | No log | 0.6087 | 28 | 1.1995 | 0.0999 | 1.1995 | 1.0952 | | No log | 0.6522 | 30 | 1.1395 | 0.1832 | 1.1395 | 1.0675 | | No log | 0.6957 | 32 | 1.1605 | 0.0436 | 1.1605 | 1.0772 | | No log | 0.7391 | 34 | 1.2550 | 0.0883 | 1.2550 | 1.1203 | | No log | 0.7826 | 36 | 1.3800 | 0.0639 | 1.3800 | 1.1747 | | No log | 0.8261 | 38 | 1.2540 | 0.0998 | 1.2540 | 1.1198 | | No log | 0.8696 | 40 | 1.1479 | 0.1389 | 1.1479 | 1.0714 | | No log | 0.9130 | 42 | 1.2217 | 0.1028 | 1.2217 | 1.1053 | | No log | 0.9565 | 44 | 1.5665 | 0.0389 | 1.5665 | 1.2516 | | No log | 1.0 | 46 | 1.6655 | 0.0516 | 1.6655 | 1.2905 | | No log | 1.0435 | 48 | 1.4103 | 0.0598 | 1.4103 | 1.1876 | | No log | 1.0870 | 50 | 1.0909 | 0.1821 | 1.0909 | 1.0444 | | No log | 1.1304 | 52 | 0.9701 | 0.3117 | 0.9701 | 0.9849 | | No log | 1.1739 | 54 | 0.9537 | 0.3414 | 0.9537 | 0.9766 | | No log | 1.2174 | 56 | 0.9618 | 0.3562 | 0.9618 | 0.9807 | | No log | 1.2609 | 58 | 0.9734 | 0.3557 | 0.9734 | 0.9866 | | No log | 1.3043 | 60 | 0.9510 | 0.3562 | 0.9510 | 0.9752 | | No log | 1.3478 | 62 | 0.9381 | 0.3733 | 0.9381 | 0.9686 | | No log | 1.3913 | 64 | 0.9339 | 0.3014 | 0.9339 | 0.9664 | | No log | 1.4348 | 66 | 0.9494 | 0.2935 | 0.9494 | 0.9743 | | No log | 1.4783 | 68 | 0.9506 | 0.2391 | 0.9506 | 0.9750 | | No log | 1.5217 | 70 | 0.9589 | 0.2114 | 0.9589 | 0.9792 | | No log | 1.5652 | 72 | 0.9002 | 0.3014 | 0.9002 | 0.9488 | | No log | 1.6087 | 74 | 0.8838 | 0.3414 | 0.8838 | 0.9401 | | No log | 1.6522 | 76 | 0.8590 | 0.3817 | 0.8590 | 0.9268 | | No log | 1.6957 | 78 | 0.8336 | 0.3519 | 0.8336 | 0.9130 | | No log | 1.7391 | 80 | 0.8431 | 0.4 | 0.8431 | 0.9182 | | No log | 1.7826 | 82 | 0.8430 | 0.3981 | 0.8430 | 0.9181 | | No log | 1.8261 | 84 | 0.8405 | 0.3537 | 0.8405 | 0.9168 | | No log | 1.8696 | 86 | 0.8561 | 0.5155 | 0.8561 | 0.9252 | | No log | 1.9130 | 88 | 0.8821 | 0.5174 | 0.8821 | 0.9392 | | No log | 1.9565 | 90 | 0.8903 | 0.4459 | 0.8903 | 0.9435 | | No log | 2.0 | 92 | 1.0073 | 0.3972 | 1.0073 | 1.0036 | | No log | 2.0435 | 94 | 1.0262 | 0.3734 | 1.0262 | 1.0130 | | No log | 2.0870 | 96 | 0.8925 | 0.5107 | 0.8925 | 0.9447 | | No log | 2.1304 | 98 | 0.8793 | 0.4979 | 0.8793 | 0.9377 | | No log | 2.1739 | 100 | 1.0192 | 0.4416 | 1.0192 | 1.0096 | | No log | 2.2174 | 102 | 0.9693 | 0.4318 | 0.9693 | 0.9845 | | No log | 2.2609 | 104 | 0.8455 | 0.5363 | 0.8455 | 0.9195 | | No log | 2.3043 | 106 | 0.8263 | 0.5450 | 0.8263 | 0.9090 | | No log | 2.3478 | 108 | 0.7896 | 0.4813 | 0.7896 | 0.8886 | | No log | 2.3913 | 110 | 0.7701 | 0.4733 | 0.7701 | 0.8776 | | No log | 2.4348 | 112 | 0.7638 | 0.4984 | 0.7638 | 0.8740 | | No log | 2.4783 | 114 | 0.7713 | 0.5128 | 0.7713 | 0.8782 | | No log | 2.5217 | 116 | 0.7180 | 0.5635 | 0.7180 | 0.8474 | | No log | 2.5652 | 118 | 0.7116 | 0.4787 | 0.7116 | 0.8436 | | No log | 2.6087 | 120 | 0.7141 | 0.5044 | 0.7141 | 0.8450 | | No log | 2.6522 | 122 | 0.7626 | 0.6386 | 0.7626 | 0.8733 | | No log | 2.6957 | 124 | 0.7743 | 0.6435 | 0.7743 | 0.8800 | | No log | 2.7391 | 126 | 0.7699 | 0.5796 | 0.7699 | 0.8774 | | No log | 2.7826 | 128 | 0.8402 | 0.4283 | 0.8402 | 0.9166 | | No log | 2.8261 | 130 | 1.0326 | 0.3363 | 1.0326 | 1.0161 | | No log | 2.8696 | 132 | 1.0287 | 0.3761 | 1.0287 | 1.0143 | | No log | 2.9130 | 134 | 0.8748 | 0.4470 | 0.8748 | 0.9353 | | No log | 2.9565 | 136 | 0.7884 | 0.5009 | 0.7884 | 0.8879 | | No log | 3.0 | 138 | 0.7676 | 0.5570 | 0.7676 | 0.8761 | | No log | 3.0435 | 140 | 0.7662 | 0.5797 | 0.7662 | 0.8753 | | No log | 3.0870 | 142 | 0.7939 | 0.5103 | 0.7939 | 0.8910 | | No log | 3.1304 | 144 | 0.8795 | 0.4455 | 0.8795 | 0.9378 | | No log | 3.1739 | 146 | 0.8895 | 0.4935 | 0.8895 | 0.9431 | | No log | 3.2174 | 148 | 0.8184 | 0.5528 | 0.8184 | 0.9046 | | No log | 3.2609 | 150 | 0.8132 | 0.4819 | 0.8132 | 0.9018 | | No log | 3.3043 | 152 | 0.8196 | 0.4615 | 0.8196 | 0.9053 | | No log | 3.3478 | 154 | 0.8178 | 0.4419 | 0.8178 | 0.9043 | | No log | 3.3913 | 156 | 0.8523 | 0.4570 | 0.8523 | 0.9232 | | No log | 3.4348 | 158 | 0.8042 | 0.4158 | 0.8042 | 0.8968 | | No log | 3.4783 | 160 | 0.8430 | 0.3160 | 0.8430 | 0.9182 | | No log | 3.5217 | 162 | 0.9556 | 0.3401 | 0.9556 | 0.9776 | | No log | 3.5652 | 164 | 0.8477 | 0.3160 | 0.8477 | 0.9207 | | No log | 3.6087 | 166 | 0.7804 | 0.5345 | 0.7804 | 0.8834 | | No log | 3.6522 | 168 | 0.8543 | 0.4806 | 0.8543 | 0.9243 | | No log | 3.6957 | 170 | 0.9146 | 0.2865 | 0.9146 | 0.9563 | | No log | 3.7391 | 172 | 0.9504 | 0.2291 | 0.9504 | 0.9749 | | No log | 3.7826 | 174 | 0.9597 | 0.2591 | 0.9597 | 0.9796 | | No log | 3.8261 | 176 | 0.9069 | 0.3445 | 0.9069 | 0.9523 | | No log | 3.8696 | 178 | 0.9335 | 0.3811 | 0.9335 | 0.9662 | | No log | 3.9130 | 180 | 0.8494 | 0.3804 | 0.8494 | 0.9216 | | No log | 3.9565 | 182 | 0.8333 | 0.4576 | 0.8333 | 0.9129 | | No log | 4.0 | 184 | 0.8048 | 0.4730 | 0.8048 | 0.8971 | | No log | 4.0435 | 186 | 0.7876 | 0.4461 | 0.7876 | 0.8875 | | No log | 4.0870 | 188 | 0.8073 | 0.3719 | 0.8073 | 0.8985 | | No log | 4.1304 | 190 | 0.7829 | 0.4128 | 0.7829 | 0.8848 | | No log | 4.1739 | 192 | 0.7779 | 0.3996 | 0.7779 | 0.8820 | | No log | 4.2174 | 194 | 0.7958 | 0.3184 | 0.7958 | 0.8921 | | No log | 4.2609 | 196 | 0.7889 | 0.3676 | 0.7889 | 0.8882 | | No log | 4.3043 | 198 | 0.7770 | 0.5163 | 0.7770 | 0.8815 | | No log | 4.3478 | 200 | 0.7898 | 0.5766 | 0.7898 | 0.8887 | | No log | 4.3913 | 202 | 0.7020 | 0.5874 | 0.7020 | 0.8378 | | No log | 4.4348 | 204 | 0.7554 | 0.5618 | 0.7554 | 0.8692 | | No log | 4.4783 | 206 | 0.7159 | 0.5379 | 0.7159 | 0.8461 | | No log | 4.5217 | 208 | 0.7211 | 0.5330 | 0.7211 | 0.8492 | | No log | 4.5652 | 210 | 0.8239 | 0.4902 | 0.8239 | 0.9077 | | No log | 4.6087 | 212 | 0.7487 | 0.5498 | 0.7487 | 0.8653 | | No log | 4.6522 | 214 | 0.7070 | 0.4898 | 0.7070 | 0.8409 | | No log | 4.6957 | 216 | 0.7224 | 0.5213 | 0.7224 | 0.8500 | | No log | 4.7391 | 218 | 0.8452 | 0.5414 | 0.8452 | 0.9194 | | No log | 4.7826 | 220 | 0.9413 | 0.4574 | 0.9413 | 0.9702 | | No log | 4.8261 | 222 | 0.8087 | 0.5231 | 0.8087 | 0.8993 | | No log | 4.8696 | 224 | 0.7436 | 0.6007 | 0.7436 | 0.8623 | | No log | 4.9130 | 226 | 0.7462 | 0.5902 | 0.7462 | 0.8638 | | No log | 4.9565 | 228 | 0.7683 | 0.5763 | 0.7683 | 0.8765 | | No log | 5.0 | 230 | 0.8623 | 0.5020 | 0.8623 | 0.9286 | | No log | 5.0435 | 232 | 0.8875 | 0.4681 | 0.8875 | 0.9421 | | No log | 5.0870 | 234 | 0.8004 | 0.5234 | 0.8004 | 0.8947 | | No log | 5.1304 | 236 | 0.7856 | 0.4691 | 0.7856 | 0.8864 | | No log | 5.1739 | 238 | 0.7886 | 0.4918 | 0.7886 | 0.8880 | | No log | 5.2174 | 240 | 0.7940 | 0.5117 | 0.7940 | 0.8911 | | No log | 5.2609 | 242 | 0.8967 | 0.4560 | 0.8967 | 0.9469 | | No log | 5.3043 | 244 | 0.9099 | 0.4987 | 0.9099 | 0.9539 | | No log | 5.3478 | 246 | 0.7790 | 0.4410 | 0.7790 | 0.8826 | | No log | 5.3913 | 248 | 0.7415 | 0.5248 | 0.7415 | 0.8611 | | No log | 5.4348 | 250 | 0.7559 | 0.4565 | 0.7559 | 0.8694 | | No log | 5.4783 | 252 | 0.7639 | 0.4261 | 0.7639 | 0.8740 | | No log | 5.5217 | 254 | 0.7792 | 0.4494 | 0.7792 | 0.8827 | | No log | 5.5652 | 256 | 0.7704 | 0.4251 | 0.7704 | 0.8777 | | No log | 5.6087 | 258 | 0.7568 | 0.4269 | 0.7568 | 0.8700 | | No log | 5.6522 | 260 | 0.7523 | 0.4313 | 0.7523 | 0.8674 | | No log | 5.6957 | 262 | 0.7222 | 0.4787 | 0.7222 | 0.8498 | | No log | 5.7391 | 264 | 0.7149 | 0.5463 | 0.7149 | 0.8455 | | No log | 5.7826 | 266 | 0.7193 | 0.6160 | 0.7193 | 0.8481 | | No log | 5.8261 | 268 | 0.7691 | 0.5439 | 0.7691 | 0.8770 | | No log | 5.8696 | 270 | 0.7113 | 0.6617 | 0.7113 | 0.8434 | | No log | 5.9130 | 272 | 0.6821 | 0.6476 | 0.6821 | 0.8259 | | No log | 5.9565 | 274 | 0.6982 | 0.6528 | 0.6982 | 0.8356 | | No log | 6.0 | 276 | 0.7793 | 0.5318 | 0.7793 | 0.8828 | | No log | 6.0435 | 278 | 0.7484 | 0.5439 | 0.7484 | 0.8651 | | No log | 6.0870 | 280 | 0.7456 | 0.5470 | 0.7456 | 0.8635 | | No log | 6.1304 | 282 | 0.7213 | 0.5740 | 0.7213 | 0.8493 | | No log | 6.1739 | 284 | 0.7244 | 0.4873 | 0.7244 | 0.8511 | | No log | 6.2174 | 286 | 0.7118 | 0.5644 | 0.7118 | 0.8437 | | No log | 6.2609 | 288 | 0.7316 | 0.4135 | 0.7316 | 0.8553 | | No log | 6.3043 | 290 | 0.7192 | 0.4984 | 0.7192 | 0.8481 | | No log | 6.3478 | 292 | 0.6941 | 0.5163 | 0.6941 | 0.8331 | | No log | 6.3913 | 294 | 0.7154 | 0.6128 | 0.7154 | 0.8458 | | No log | 6.4348 | 296 | 0.7172 | 0.5618 | 0.7172 | 0.8469 | | No log | 6.4783 | 298 | 0.6939 | 0.5680 | 0.6939 | 0.8330 | | No log | 6.5217 | 300 | 0.6980 | 0.5877 | 0.6980 | 0.8354 | | No log | 6.5652 | 302 | 0.7417 | 0.5052 | 0.7417 | 0.8612 | | No log | 6.6087 | 304 | 0.7484 | 0.5067 | 0.7484 | 0.8651 | | No log | 6.6522 | 306 | 0.7230 | 0.4879 | 0.7230 | 0.8503 | | No log | 6.6957 | 308 | 0.7159 | 0.5002 | 0.7159 | 0.8461 | | No log | 6.7391 | 310 | 0.7043 | 0.5357 | 0.7043 | 0.8392 | | No log | 6.7826 | 312 | 0.6960 | 0.5060 | 0.6960 | 0.8343 | | No log | 6.8261 | 314 | 0.6992 | 0.5066 | 0.6992 | 0.8362 | | No log | 6.8696 | 316 | 0.6768 | 0.5809 | 0.6768 | 0.8227 | | No log | 6.9130 | 318 | 0.6808 | 0.5329 | 0.6808 | 0.8251 | | No log | 6.9565 | 320 | 0.6908 | 0.5671 | 0.6908 | 0.8312 | | No log | 7.0 | 322 | 0.7242 | 0.5263 | 0.7242 | 0.8510 | | No log | 7.0435 | 324 | 0.7106 | 0.5459 | 0.7106 | 0.8430 | | No log | 7.0870 | 326 | 0.8062 | 0.4708 | 0.8062 | 0.8979 | | No log | 7.1304 | 328 | 0.8057 | 0.4708 | 0.8057 | 0.8976 | | No log | 7.1739 | 330 | 0.7553 | 0.5046 | 0.7553 | 0.8691 | | No log | 7.2174 | 332 | 0.7268 | 0.5094 | 0.7268 | 0.8525 | | No log | 7.2609 | 334 | 0.7145 | 0.5582 | 0.7145 | 0.8453 | | No log | 7.3043 | 336 | 0.7106 | 0.5475 | 0.7106 | 0.8430 | | No log | 7.3478 | 338 | 0.7219 | 0.5446 | 0.7219 | 0.8497 | | No log | 7.3913 | 340 | 0.7267 | 0.5129 | 0.7267 | 0.8525 | | No log | 7.4348 | 342 | 0.7339 | 0.4461 | 0.7339 | 0.8567 | | No log | 7.4783 | 344 | 0.7684 | 0.4641 | 0.7684 | 0.8766 | | No log | 7.5217 | 346 | 0.7471 | 0.4641 | 0.7471 | 0.8644 | | No log | 7.5652 | 348 | 0.7191 | 0.5352 | 0.7191 | 0.8480 | | No log | 7.6087 | 350 | 0.7407 | 0.5953 | 0.7407 | 0.8606 | | No log | 7.6522 | 352 | 0.7185 | 0.5683 | 0.7185 | 0.8477 | | No log | 7.6957 | 354 | 0.7478 | 0.5476 | 0.7478 | 0.8647 | | No log | 7.7391 | 356 | 0.7586 | 0.5197 | 0.7586 | 0.8710 | | No log | 7.7826 | 358 | 0.7467 | 0.4996 | 0.7467 | 0.8641 | | No log | 7.8261 | 360 | 0.8009 | 0.4388 | 0.8009 | 0.8949 | | No log | 7.8696 | 362 | 0.8052 | 0.3269 | 0.8052 | 0.8973 | | No log | 7.9130 | 364 | 0.7747 | 0.4279 | 0.7747 | 0.8802 | | No log | 7.9565 | 366 | 0.7503 | 0.4660 | 0.7503 | 0.8662 | | No log | 8.0 | 368 | 0.7182 | 0.4918 | 0.7182 | 0.8475 | | No log | 8.0435 | 370 | 0.6882 | 0.5120 | 0.6882 | 0.8296 | | No log | 8.0870 | 372 | 0.6722 | 0.6187 | 0.6722 | 0.8199 | | No log | 8.1304 | 374 | 0.7175 | 0.6081 | 0.7175 | 0.8470 | | No log | 8.1739 | 376 | 0.7358 | 0.6071 | 0.7358 | 0.8578 | | No log | 8.2174 | 378 | 0.7982 | 0.5398 | 0.7982 | 0.8934 | | No log | 8.2609 | 380 | 0.7700 | 0.6218 | 0.7700 | 0.8775 | | No log | 8.3043 | 382 | 0.6887 | 0.5463 | 0.6887 | 0.8299 | | No log | 8.3478 | 384 | 0.8329 | 0.4508 | 0.8329 | 0.9127 | | No log | 8.3913 | 386 | 0.9849 | 0.5184 | 0.9849 | 0.9924 | | No log | 8.4348 | 388 | 0.9149 | 0.4854 | 0.9149 | 0.9565 | | No log | 8.4783 | 390 | 0.7384 | 0.5012 | 0.7384 | 0.8593 | | No log | 8.5217 | 392 | 0.7104 | 0.5473 | 0.7104 | 0.8428 | | No log | 8.5652 | 394 | 0.8558 | 0.4894 | 0.8558 | 0.9251 | | No log | 8.6087 | 396 | 0.8252 | 0.4894 | 0.8252 | 0.9084 | | No log | 8.6522 | 398 | 0.7126 | 0.5186 | 0.7126 | 0.8441 | | No log | 8.6957 | 400 | 0.6932 | 0.5432 | 0.6932 | 0.8326 | | No log | 8.7391 | 402 | 0.7075 | 0.4968 | 0.7075 | 0.8412 | | No log | 8.7826 | 404 | 0.7033 | 0.5432 | 0.7033 | 0.8387 | | No log | 8.8261 | 406 | 0.7082 | 0.5536 | 0.7082 | 0.8415 | | No log | 8.8696 | 408 | 0.7080 | 0.4760 | 0.7080 | 0.8414 | | No log | 8.9130 | 410 | 0.6967 | 0.4760 | 0.6967 | 0.8347 | | No log | 8.9565 | 412 | 0.6681 | 0.5822 | 0.6681 | 0.8174 | | No log | 9.0 | 414 | 0.6506 | 0.6046 | 0.6506 | 0.8066 | | No log | 9.0435 | 416 | 0.6532 | 0.6219 | 0.6532 | 0.8082 | | No log | 9.0870 | 418 | 0.6514 | 0.6219 | 0.6514 | 0.8071 | | No log | 9.1304 | 420 | 0.6505 | 0.5644 | 0.6505 | 0.8066 | | No log | 9.1739 | 422 | 0.6544 | 0.5886 | 0.6544 | 0.8090 | | No log | 9.2174 | 424 | 0.6797 | 0.5597 | 0.6797 | 0.8245 | | No log | 9.2609 | 426 | 0.7490 | 0.5137 | 0.7490 | 0.8655 | | No log | 9.3043 | 428 | 0.7794 | 0.4898 | 0.7794 | 0.8828 | | No log | 9.3478 | 430 | 0.7201 | 0.5400 | 0.7201 | 0.8486 | | No log | 9.3913 | 432 | 0.6806 | 0.5432 | 0.6806 | 0.8250 | | No log | 9.4348 | 434 | 0.6772 | 0.5432 | 0.6772 | 0.8229 | | No log | 9.4783 | 436 | 0.6891 | 0.5089 | 0.6891 | 0.8301 | | No log | 9.5217 | 438 | 0.7677 | 0.5428 | 0.7677 | 0.8762 | | No log | 9.5652 | 440 | 0.7609 | 0.5451 | 0.7609 | 0.8723 | | No log | 9.6087 | 442 | 0.6901 | 0.5570 | 0.6901 | 0.8307 | | No log | 9.6522 | 444 | 0.6739 | 0.5188 | 0.6739 | 0.8209 | | No log | 9.6957 | 446 | 0.6775 | 0.5074 | 0.6775 | 0.8231 | | No log | 9.7391 | 448 | 0.6750 | 0.5516 | 0.6750 | 0.8216 | | No log | 9.7826 | 450 | 0.7319 | 0.5279 | 0.7319 | 0.8555 | | No log | 9.8261 | 452 | 0.7593 | 0.5331 | 0.7593 | 0.8714 | | No log | 9.8696 | 454 | 0.6942 | 0.5098 | 0.6942 | 0.8332 | | No log | 9.9130 | 456 | 0.6676 | 0.5771 | 0.6676 | 0.8171 | | No log | 9.9565 | 458 | 0.6685 | 0.5783 | 0.6685 | 0.8176 | | No log | 10.0 | 460 | 0.6652 | 0.6307 | 0.6652 | 0.8156 | | No log | 10.0435 | 462 | 0.7046 | 0.5395 | 0.7046 | 0.8394 | | No log | 10.0870 | 464 | 0.7331 | 0.5470 | 0.7331 | 0.8562 | | No log | 10.1304 | 466 | 0.7124 | 0.4494 | 0.7124 | 0.8440 | | No log | 10.1739 | 468 | 0.7124 | 0.4893 | 0.7124 | 0.8441 | | No log | 10.2174 | 470 | 0.7132 | 0.4923 | 0.7132 | 0.8445 | | No log | 10.2609 | 472 | 0.7016 | 0.5415 | 0.7016 | 0.8376 | | No log | 10.3043 | 474 | 0.6915 | 0.5554 | 0.6915 | 0.8316 | | No log | 10.3478 | 476 | 0.6783 | 0.6177 | 0.6783 | 0.8236 | | No log | 10.3913 | 478 | 0.7239 | 0.5862 | 0.7239 | 0.8508 | | No log | 10.4348 | 480 | 0.7312 | 0.5958 | 0.7312 | 0.8551 | | No log | 10.4783 | 482 | 0.6749 | 0.6198 | 0.6749 | 0.8215 | | No log | 10.5217 | 484 | 0.6417 | 0.6335 | 0.6417 | 0.8011 | | No log | 10.5652 | 486 | 0.6504 | 0.5084 | 0.6504 | 0.8065 | | No log | 10.6087 | 488 | 0.6547 | 0.4968 | 0.6547 | 0.8091 | | No log | 10.6522 | 490 | 0.6450 | 0.5529 | 0.6450 | 0.8031 | | No log | 10.6957 | 492 | 0.6673 | 0.6073 | 0.6673 | 0.8169 | | No log | 10.7391 | 494 | 0.6924 | 0.5973 | 0.6924 | 0.8321 | | No log | 10.7826 | 496 | 0.6905 | 0.6109 | 0.6905 | 0.8310 | | No log | 10.8261 | 498 | 0.6965 | 0.5833 | 0.6965 | 0.8346 | | 0.2902 | 10.8696 | 500 | 0.7077 | 0.6209 | 0.7077 | 0.8412 | | 0.2902 | 10.9130 | 502 | 0.6775 | 0.6147 | 0.6775 | 0.8231 | | 0.2902 | 10.9565 | 504 | 0.6453 | 0.6500 | 0.6453 | 0.8033 | | 0.2902 | 11.0 | 506 | 0.6393 | 0.6175 | 0.6393 | 0.7996 | | 0.2902 | 11.0435 | 508 | 0.6427 | 0.6057 | 0.6427 | 0.8017 | | 0.2902 | 11.0870 | 510 | 0.6740 | 0.5932 | 0.6740 | 0.8209 | | 0.2902 | 11.1304 | 512 | 0.7058 | 0.5654 | 0.7058 | 0.8401 | | 0.2902 | 11.1739 | 514 | 0.6862 | 0.5472 | 0.6862 | 0.8284 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.0+cu118 - Datasets 2.21.0 - Tokenizers 0.19.1

mrhunghd/3ec0da79-0eab-4cb1-a6c3-1edba33530b8

mrhunghd

2025-01-20T23:02:42Z

6

0

peft

[ "peft", "safetensors", "opt", "axolotl", "generated_from_trainer", "base_model:facebook/opt-125m", "base_model:adapter:facebook/opt-125m", "license:other", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:58:34Z

--- library_name: peft license: other base_model: facebook/opt-125m tags: - axolotl - generated_from_trainer model-index: - name: 3ec0da79-0eab-4cb1-a6c3-1edba33530b8 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: facebook/opt-125m bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 1b096fb12091d0a7_train_data.json ds_type: json format: custom path: /workspace/input_data/1b096fb12091d0a7_train_data.json type: field_instruction: problem field_output: solution format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: mrhunghd/3ec0da79-0eab-4cb1-a6c3-1edba33530b8 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/1b096fb12091d0a7_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 2594ef14-2fe6-455c-8347-c1d0fb26863f wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 2594ef14-2fe6-455c-8347-c1d0fb26863f warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 3ec0da79-0eab-4cb1-a6c3-1edba33530b8 This model is a fine-tuned version of [facebook/opt-125m](https://huggingface.co/facebook/opt-125m) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.1761 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 9.3655 | 0.2159 | 200 | 2.1761 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

nblinh63/f1b226a4-cdd1-4308-ac2d-1fa4ebe01843

nblinh63

2025-01-20T23:02:37Z

6

0

peft

[ "peft", "safetensors", "opt", "axolotl", "generated_from_trainer", "base_model:facebook/opt-125m", "base_model:adapter:facebook/opt-125m", "license:other", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:58:31Z

--- library_name: peft license: other base_model: facebook/opt-125m tags: - axolotl - generated_from_trainer model-index: - name: f1b226a4-cdd1-4308-ac2d-1fa4ebe01843 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: facebook/opt-125m bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 1b096fb12091d0a7_train_data.json ds_type: json format: custom path: /workspace/input_data/1b096fb12091d0a7_train_data.json type: field_instruction: problem field_output: solution format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nblinh63/f1b226a4-cdd1-4308-ac2d-1fa4ebe01843 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/1b096fb12091d0a7_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 2594ef14-2fe6-455c-8347-c1d0fb26863f wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 2594ef14-2fe6-455c-8347-c1d0fb26863f warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # f1b226a4-cdd1-4308-ac2d-1fa4ebe01843 This model is a fine-tuned version of [facebook/opt-125m](https://huggingface.co/facebook/opt-125m) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.1806 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 9.3858 | 0.2159 | 200 | 2.1806 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

dzanbek/f58f03f7-8fc0-41ef-9d07-5565794dc71c

dzanbek

2025-01-20T23:02:31Z

9

0

peft

[ "peft", "safetensors", "qwen2", "axolotl", "generated_from_trainer", "base_model:unsloth/Qwen2-7B-Instruct", "base_model:adapter:unsloth/Qwen2-7B-Instruct", "license:apache-2.0", "region:us" ]

null

2025-01-20T22:22:46Z

--- library_name: peft license: apache-2.0 base_model: unsloth/Qwen2-7B-Instruct tags: - axolotl - generated_from_trainer model-index: - name: f58f03f7-8fc0-41ef-9d07-5565794dc71c results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: unsloth/Qwen2-7B-Instruct bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 6e60a538f672529c_train_data.json ds_type: json format: custom path: /workspace/input_data/6e60a538f672529c_train_data.json type: field_input: communityName field_instruction: label field_output: text format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: false hub_model_id: dzanbek/f58f03f7-8fc0-41ef-9d07-5565794dc71c hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 78GiB max_steps: 30 micro_batch_size: 2 mlflow_experiment_name: /tmp/6e60a538f672529c_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 3eb92360-4e77-4cc9-9ffa-0e03d7ea7423 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 3eb92360-4e77-4cc9-9ffa-0e03d7ea7423 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # f58f03f7-8fc0-41ef-9d07-5565794dc71c This model is a fine-tuned version of [unsloth/Qwen2-7B-Instruct](https://huggingface.co/unsloth/Qwen2-7B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0001 | 1 | nan | | 0.0 | 0.0004 | 5 | nan | | 0.0 | 0.0008 | 10 | nan | | 0.0 | 0.0012 | 15 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

mrHungddddh/0ff89366-d444-4953-ad52-fc6a503d92cb

mrHungddddh

2025-01-20T23:01:12Z

6

0

peft

[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/Yarn-Mistral-7b-64k", "base_model:adapter:NousResearch/Yarn-Mistral-7b-64k", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T21:58:35Z

--- library_name: peft license: apache-2.0 base_model: NousResearch/Yarn-Mistral-7b-64k tags: - axolotl - generated_from_trainer model-index: - name: 0ff89366-d444-4953-ad52-fc6a503d92cb results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/Yarn-Mistral-7b-64k bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 5f5f73d0b6f6fe1d_train_data.json ds_type: json format: custom path: /workspace/input_data/5f5f73d0b6f6fe1d_train_data.json type: field_input: messages field_instruction: system field_output: reference format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: mrHungddddh/0ff89366-d444-4953-ad52-fc6a503d92cb hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/5f5f73d0b6f6fe1d_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 6a440f8b-4ef2-40c5-aaae-529ca715837e wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 6a440f8b-4ef2-40c5-aaae-529ca715837e warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 0ff89366-d444-4953-ad52-fc6a503d92cb This model is a fine-tuned version of [NousResearch/Yarn-Mistral-7b-64k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-64k) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.4651 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 5.1264 | 0.0263 | 200 | 1.4651 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

VERSIL91/b9a1e459-4362-4834-8b97-f53708a182cd

VERSIL91

2025-01-20T23:00:32Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:huggyllama/llama-7b", "base_model:adapter:huggyllama/llama-7b", "license:other", "region:us" ]

null

2025-01-20T22:53:20Z

--- library_name: peft license: other base_model: huggyllama/llama-7b tags: - axolotl - generated_from_trainer model-index: - name: b9a1e459-4362-4834-8b97-f53708a182cd results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml accelerate_config: dynamo_backend: inductor mixed_precision: bf16 num_machines: 1 num_processes: auto use_cpu: false adapter: lora base_model: huggyllama/llama-7b bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 804c65db1d2351c5_train_data.json ds_type: json format: custom path: /workspace/input_data/804c65db1d2351c5_train_data.json type: field_instruction: sentence1 field_output: sentence2 format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 4 flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 16 gradient_checkpointing: true group_by_length: false hub_model_id: VERSIL91/b9a1e459-4362-4834-8b97-f53708a182cd hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lora_target_modules: - q_proj - v_proj lr_scheduler: cosine max_memory: 0: 70GiB max_steps: 50 micro_batch_size: 2 mlflow_experiment_name: /tmp/804c65db1d2351c5_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true quantization_config: llm_int8_enable_fp32_cpu_offload: true load_in_8bit: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 4 sequence_len: 512 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer torch_compile: true train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: b9a1e459-4362-4834-8b97-f53708a182cd wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: b9a1e459-4362-4834-8b97-f53708a182cd warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # b9a1e459-4362-4834-8b97-f53708a182cd This model is a fine-tuned version of [huggyllama/llama-7b](https://huggingface.co/huggyllama/llama-7b) on the None dataset. It achieves the following results on the evaluation set: - Loss: nan ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 16 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 50 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 0.0 | 0.0010 | 1 | nan | | 0.0 | 0.0126 | 13 | nan | | 0.0 | 0.0252 | 26 | nan | | 0.0 | 0.0377 | 39 | nan | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

mradermacher/FastApply-32B-Instruct-GGUF

mradermacher

2025-01-20T23:00:06Z

302

0

transformers

[ "transformers", "gguf", "text-generation-inference", "unsloth", "qwen2", "trl", "en", "base_model:tabnine/FastApply-32B-Instruct", "base_model:quantized:tabnine/FastApply-32B-Instruct", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]

null

2025-01-20T20:47:56Z

--- base_model: tabnine/FastApply-32B-Instruct language: - en library_name: transformers license: apache-2.0 quantized_by: mradermacher tags: - text-generation-inference - transformers - unsloth - qwen2 - trl --- ## About      static quants of https://huggingface.co/tabnine/FastApply-32B-Instruct  weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q2_K.gguf) | Q2_K | 12.4 | | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q3_K_S.gguf) | Q3_K_S | 14.5 | | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q3_K_M.gguf) | Q3_K_M | 16.0 | lower quality | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q3_K_L.gguf) | Q3_K_L | 17.3 | | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.IQ4_XS.gguf) | IQ4_XS | 18.0 | | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q4_K_S.gguf) | Q4_K_S | 18.9 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q4_K_M.gguf) | Q4_K_M | 20.0 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q5_K_S.gguf) | Q5_K_S | 22.7 | | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q5_K_M.gguf) | Q5_K_M | 23.4 | | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q6_K.gguf) | Q6_K | 27.0 | very good quality | | [GGUF](https://huggingface.co/mradermacher/FastApply-32B-Instruct-GGUF/resolve/main/FastApply-32B-Instruct.Q8_0.gguf) | Q8_0 | 34.9 | fast, best quality | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time.

nhunglaaaaaaa/31738ab5-8e3e-4c3d-a223-8d795b64609e

nhunglaaaaaaa

2025-01-20T22:58:35Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:heegyu/WizardVicuna-open-llama-3b-v2", "base_model:adapter:heegyu/WizardVicuna-open-llama-3b-v2", "license:apache-2.0", "8-bit", "bitsandbytes", "region:us" ]

null

2025-01-20T22:05:54Z

--- library_name: peft license: apache-2.0 base_model: heegyu/WizardVicuna-open-llama-3b-v2 tags: - axolotl - generated_from_trainer model-index: - name: 31738ab5-8e3e-4c3d-a223-8d795b64609e results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: heegyu/WizardVicuna-open-llama-3b-v2 bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 212d5e0168a48c19_train_data.json ds_type: json format: custom path: /workspace/input_data/212d5e0168a48c19_train_data.json type: field_instruction: context_en field_output: question_en format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true gradient_clipping: 1.0 group_by_length: false hub_model_id: nhunglaaaaaaa/31738ab5-8e3e-4c3d-a223-8d795b64609e hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-05 load_in_4bit: true load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 16 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 8 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 2 mlflow_experiment_name: /tmp/212d5e0168a48c19_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 1e1fd096-ba6e-478c-9e4b-c08c22fc3c74 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 1e1fd096-ba6e-478c-9e4b-c08c22fc3c74 warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 31738ab5-8e3e-4c3d-a223-8d795b64609e This model is a fine-tuned version of [heegyu/WizardVicuna-open-llama-3b-v2](https://huggingface.co/heegyu/WizardVicuna-open-llama-3b-v2) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.8120 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.0691 | 0.0026 | 200 | 0.8120 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

dimasik1987/0b3322d1-4a39-4579-a195-f5358131b723

dimasik1987

2025-01-20T22:58:05Z

6

0

peft

[ "peft", "safetensors", "mistral", "axolotl", "generated_from_trainer", "custom_code", "base_model:NousResearch/Yarn-Mistral-7b-64k", "base_model:adapter:NousResearch/Yarn-Mistral-7b-64k", "license:apache-2.0", "region:us" ]

null

2025-01-20T21:58:35Z

--- library_name: peft license: apache-2.0 base_model: NousResearch/Yarn-Mistral-7b-64k tags: - axolotl - generated_from_trainer model-index: - name: 0b3322d1-4a39-4579-a195-f5358131b723 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: NousResearch/Yarn-Mistral-7b-64k bf16: auto chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - 5f5f73d0b6f6fe1d_train_data.json ds_type: json format: custom path: /workspace/input_data/5f5f73d0b6f6fe1d_train_data.json type: field_input: messages field_instruction: system field_output: reference format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device: cuda early_stopping_patience: 1 eval_max_new_tokens: 128 eval_steps: 5 eval_table_size: null evals_per_epoch: null flash_attention: false fp16: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: dimasik1987/0b3322d1-4a39-4579-a195-f5358131b723 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0002 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 3 lora_alpha: 32 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 16 lora_target_linear: true lr_scheduler: cosine max_memory: 0: 79GiB max_steps: 30 micro_batch_size: 4 mlflow_experiment_name: /tmp/5f5f73d0b6f6fe1d_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 10 sequence_len: 1024 special_tokens: pad_token: </s> strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: true trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 6a440f8b-4ef2-40c5-aaae-529ca715837e wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 6a440f8b-4ef2-40c5-aaae-529ca715837e warmup_steps: 5 weight_decay: 0.001 xformers_attention: true ``` </details><br> # 0b3322d1-4a39-4579-a195-f5358131b723 This model is a fine-tuned version of [NousResearch/Yarn-Mistral-7b-64k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-64k) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.8223 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 30 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.0003 | 1 | 1.2169 | | 3.8291 | 0.0013 | 5 | 1.0084 | | 3.1617 | 0.0026 | 10 | 0.8943 | | 3.1678 | 0.0039 | 15 | 0.8525 | | 3.2285 | 0.0053 | 20 | 0.8367 | | 3.3153 | 0.0066 | 25 | 0.8247 | | 3.5174 | 0.0079 | 30 | 0.8223 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1

lilmeaty/xfsfsfsf-4bit

lilmeaty

2025-01-20T22:57:54Z

20

0

transformers

[ "transformers", "pytorch", "safetensors", "llama", "text-generation", "conversational", "license:apache-2.0", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]

text-generation

2025-01-20T07:50:13Z

--- license: apache-2.0 library_name: transformers ---

prxy5607/feb3717a-696e-4e00-8c82-54646ee762f9

prxy5607

2025-01-20T22:56:12Z

6

0

peft

[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:deepseek-ai/deepseek-coder-6.7b-instruct", "base_model:adapter:deepseek-ai/deepseek-coder-6.7b-instruct", "license:other", "region:us" ]

null

2025-01-20T21:36:32Z

--- library_name: peft license: other base_model: deepseek-ai/deepseek-coder-6.7b-instruct tags: - axolotl - generated_from_trainer model-index: - name: feb3717a-696e-4e00-8c82-54646ee762f9 results: [] ---  [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: deepseek-ai/deepseek-coder-6.7b-instruct bf16: true chat_template: llama3 data_processes: 16 dataset_prepared_path: null datasets: - data_files: - aaf4cd02348b6ba9_train_data.json ds_type: json format: custom path: /workspace/input_data/aaf4cd02348b6ba9_train_data.json type: field_instruction: code field_output: docstring format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: auto do_eval: true early_stopping_patience: 5 eval_batch_size: 4 eval_max_new_tokens: 128 eval_steps: 50 eval_table_size: null evals_per_epoch: null flash_attention: true fp16: false fsdp: null fsdp_config: null gradient_accumulation_steps: 4 gradient_checkpointing: true group_by_length: true hub_model_id: prxy5607/feb3717a-696e-4e00-8c82-54646ee762f9 hub_repo: null hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 128 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 64 lora_target_linear: true lr_scheduler: cosine max_grad_norm: 1.0 max_memory: 0: 75GB max_steps: 200 micro_batch_size: 8 mlflow_experiment_name: /tmp/aaf4cd02348b6ba9_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 50 saves_per_epoch: null sequence_len: 1024 strict: false tf32: true tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 6b80479d-4e03-4f5b-b68e-f811da024a88 wandb_project: Gradients-On-Demand wandb_run: your_name wandb_runid: 6b80479d-4e03-4f5b-b68e-f811da024a88 warmup_steps: 10 weight_decay: 0.0 xformers_attention: null ``` </details><br> # feb3717a-696e-4e00-8c82-54646ee762f9 This model is a fine-tuned version of [deepseek-ai/deepseek-coder-6.7b-instruct](https://huggingface.co/deepseek-ai/deepseek-coder-6.7b-instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2623 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 8 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 10 - training_steps: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 2.6902 | 0.0002 | 1 | 3.2009 | | 1.0728 | 0.0092 | 50 | 1.3635 | | 1.2867 | 0.0185 | 100 | 1.2835 | | 1.3301 | 0.0277 | 150 | 1.2678 | | 1.3324 | 0.0369 | 200 | 1.2623 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1