diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..133d8bc2e1b8357650ff474ea715e475204a1e0f --- /dev/null +++ b/README.md @@ -0,0 +1,143 @@ +--- +license: apache-2.0 +library_name: peft +tags: +- generated_from_trainer +base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T +model-index: +- name: outputs/qlora-out + results: [] +--- + + + +[Built with Axolotl](https://github.com/OpenAccess-AI-Collective/axolotl) +
See axolotl config + +axolotl version: `0.4.1` +```yaml +adapter: qlora +base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T +bf16: false +dataset_prepared_path: null +datasets: +- path: mhenrichsen/alpaca_2k_test + type: alpaca +debug: null +deepspeed: null +early_stopping_patience: null +eval_sample_packing: false +evals_per_epoch: 4 +flash_attention: false +fp16: null +fsdp: null +fsdp_config: null +gradient_accumulation_steps: 4 +gradient_checkpointing: true +group_by_length: false +learning_rate: 0.0002 +load_in_4bit: true +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: 32 +lora_target_linear: true +lora_target_modules: null +lr_scheduler: cosine +micro_batch_size: 2 +model_type: LlamaForCausalLM +num_epochs: 4 +optimizer: paged_adamw_32bit +output_dir: ./outputs/qlora-out +pad_to_sequence_len: false +resume_from_checkpoint: null +sample_packing: false +saves_per_epoch: 1 +sequence_len: 4096 +special_tokens: null +strict: false +tf32: null +tokenizer_type: LlamaTokenizer +train_on_inputs: false +val_set_size: 0.05 +wandb_entity: null +wandb_log_model: null +wandb_name: null +wandb_project: null +wandb_watch: null +warmup_steps: 10 +weight_decay: 0.0 +xformers_attention: null + +``` + +

+ +# outputs/qlora-out + +This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the None dataset. +It achieves the following results on the evaluation set: +- Loss: 1.2295 + +## 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: Adam with betas=(0.9,0.999) and epsilon=1e-08 +- lr_scheduler_type: cosine +- lr_scheduler_warmup_steps: 10 +- num_epochs: 4 + +### Training results + +| Training Loss | Epoch | Step | Validation Loss | +|:-------------:|:------:|:----:|:---------------:| +| 1.5605 | 0.0042 | 1 | 1.5265 | +| 1.1485 | 0.2526 | 60 | 1.2386 | +| 1.1249 | 0.5053 | 120 | 1.2167 | +| 1.3675 | 0.7579 | 180 | 1.2130 | +| 1.3449 | 1.0105 | 240 | 1.1995 | +| 1.1825 | 1.2632 | 300 | 1.2074 | +| 0.9782 | 1.5158 | 360 | 1.2060 | +| 1.2063 | 1.7684 | 420 | 1.1994 | +| 0.9614 | 2.0211 | 480 | 1.1929 | +| 1.0084 | 2.2737 | 540 | 1.2140 | +| 1.1655 | 2.5263 | 600 | 1.2174 | +| 1.1503 | 2.7789 | 660 | 1.2198 | +| 0.9577 | 3.0316 | 720 | 1.2164 | +| 0.9943 | 3.2842 | 780 | 1.2286 | +| 1.0043 | 3.5368 | 840 | 1.2289 | +| 1.128 | 3.7895 | 900 | 1.2295 | + + +### Framework versions + +- PEFT 0.11.1 +- Transformers 4.41.1 +- Pytorch 2.1.2+cu121 +- Datasets 2.19.1 +- Tokenizers 0.19.1 \ No newline at end of file diff --git a/adapter_config.json b/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..cfa45a23cb958fe6ad106d4adf7f6169698c1fc8 --- /dev/null +++ b/adapter_config.json @@ -0,0 +1,34 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 16, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "v_proj", + "k_proj", + "q_proj", + "gate_proj", + "up_proj", + "down_proj", + "o_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/adapter_model.bin b/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..7abfbec83abfd3b0a9c55c00d73cc1bbd0ef4bd9 --- /dev/null +++ b/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fe890ad8ee220f505bc4a0ebd66cf525a3d15a760985d3af21a1e0e8212e40b4 +size 50573978 diff --git a/checkpoint-237/README.md b/checkpoint-237/README.md new file mode 100644 index 0000000000000000000000000000000000000000..e1ccd431539a8f1507d8755a9c3ba5e5b2897978 --- /dev/null +++ b/checkpoint-237/README.md @@ -0,0 +1,202 @@ +--- +library_name: peft +base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **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] +### Framework versions + +- PEFT 0.11.1 \ No newline at end of file diff --git a/checkpoint-237/adapter_config.json b/checkpoint-237/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..cfa45a23cb958fe6ad106d4adf7f6169698c1fc8 --- /dev/null +++ b/checkpoint-237/adapter_config.json @@ -0,0 +1,34 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 16, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "v_proj", + "k_proj", + "q_proj", + "gate_proj", + "up_proj", + "down_proj", + "o_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-237/adapter_model.safetensors b/checkpoint-237/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..903a690e547ffc79d2fc96cfbf8de3a807c2a186 --- /dev/null +++ b/checkpoint-237/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c0117ce11d9f2e5d6bd5f133ad30669999be08dbaa426121bbd98337975bf835 +size 50503848 diff --git a/checkpoint-237/optimizer.pt b/checkpoint-237/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..4a8184b3ad4cecb0662ae19d123541587dbe9b41 --- /dev/null +++ b/checkpoint-237/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e8acf16b4e1bd9dc6c260276356388e91496c718a307a28cecd8e4a79af1e2f +size 202035450 diff --git a/checkpoint-237/rng_state.pth b/checkpoint-237/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..0c63b0484ec4b861f8c9814a1e7692502fe5b4d6 --- /dev/null +++ b/checkpoint-237/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1746df72a4bad5eddb1e3c7cb25ea26a49bf4140a3ca0312fff3900f3c96776 +size 14244 diff --git a/checkpoint-237/scheduler.pt b/checkpoint-237/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..c3c069dd5c4566e0ec5f441b4bcdc44746afd5d0 --- /dev/null +++ b/checkpoint-237/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9d93959d5a6796fe7eb3f081e11fa0927ee021d92c2af05266234d62643f7e6c +size 1064 diff --git a/checkpoint-237/special_tokens_map.json b/checkpoint-237/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1 --- /dev/null +++ b/checkpoint-237/special_tokens_map.json @@ -0,0 +1,24 @@ +{ + "bos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "", + "unk_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-237/tokenizer.model b/checkpoint-237/tokenizer.model new file mode 100644 index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899 --- /dev/null +++ b/checkpoint-237/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 +size 499723 diff --git a/checkpoint-237/tokenizer_config.json b/checkpoint-237/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..0773857a13ba5a27453a0b462624fe76e8e82a86 --- /dev/null +++ b/checkpoint-237/tokenizer_config.json @@ -0,0 +1,44 @@ +{ + "add_bos_token": true, + "add_eos_token": false, + "add_prefix_space": true, + "added_tokens_decoder": { + "0": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "1": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "2": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "", + "clean_up_tokenization_spaces": false, + "eos_token": "", + "legacy": false, + "model_max_length": 1000000000000000019884624838656, + "pad_token": "", + "padding_side": "right", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "LlamaTokenizer", + "unk_token": "", + "use_default_system_prompt": false, + "use_fast": true +} diff --git a/checkpoint-237/trainer_state.json b/checkpoint-237/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..0e37498b2c73533f7b5d4d089b59de21c1cc8529 --- /dev/null +++ b/checkpoint-237/trainer_state.json @@ -0,0 +1,1724 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.9978947368421053, + "eval_steps": 60, + "global_step": 237, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.004210526315789474, + "grad_norm": 0.310546875, + "learning_rate": 2e-05, + "loss": 1.5605, + "step": 1 + }, + { + "epoch": 0.004210526315789474, + "eval_loss": 1.5264556407928467, + "eval_runtime": 26.3082, + "eval_samples_per_second": 3.801, + "eval_steps_per_second": 1.901, + "step": 1 + }, + { + "epoch": 0.008421052631578947, + "grad_norm": 0.306640625, + "learning_rate": 4e-05, + "loss": 1.4702, + "step": 2 + }, + { + "epoch": 0.01263157894736842, + "grad_norm": 0.2578125, + "learning_rate": 6e-05, + "loss": 1.4576, + "step": 3 + }, + { + "epoch": 0.016842105263157894, + "grad_norm": 0.65234375, + "learning_rate": 8e-05, + "loss": 1.5868, + "step": 4 + }, + { + "epoch": 0.021052631578947368, + "grad_norm": 0.56640625, + "learning_rate": 0.0001, + "loss": 1.5916, + "step": 5 + }, + { + "epoch": 0.02526315789473684, + "grad_norm": 0.5703125, + "learning_rate": 0.00012, + "loss": 1.7851, + "step": 6 + }, + { + "epoch": 0.029473684210526315, + "grad_norm": 0.3359375, + "learning_rate": 0.00014, + "loss": 1.5319, + "step": 7 + }, + { + "epoch": 0.03368421052631579, + "grad_norm": 0.39453125, + "learning_rate": 0.00016, + "loss": 1.8462, + "step": 8 + }, + { + "epoch": 0.037894736842105266, + "grad_norm": 0.484375, + "learning_rate": 0.00018, + "loss": 1.4219, + "step": 9 + }, + { + "epoch": 0.042105263157894736, + "grad_norm": 0.30859375, + "learning_rate": 0.0002, + "loss": 1.6555, + "step": 10 + }, + { + "epoch": 0.04631578947368421, + "grad_norm": 0.267578125, + "learning_rate": 0.0001999994391281196, + "loss": 1.1527, + "step": 11 + }, + { + "epoch": 0.05052631578947368, + "grad_norm": 0.298828125, + "learning_rate": 0.00019999775651876987, + "loss": 1.4265, + "step": 12 + }, + { + "epoch": 0.05473684210526316, + "grad_norm": 0.376953125, + "learning_rate": 0.00019999495219082545, + "loss": 1.5505, + "step": 13 + }, + { + "epoch": 0.05894736842105263, + "grad_norm": 0.380859375, + "learning_rate": 0.00019999102617574365, + "loss": 1.0313, + "step": 14 + }, + { + "epoch": 0.06315789473684211, + "grad_norm": 0.494140625, + "learning_rate": 0.0001999859785175643, + "loss": 1.439, + "step": 15 + }, + { + "epoch": 0.06736842105263158, + "grad_norm": 0.40234375, + "learning_rate": 0.00019997980927290927, + "loss": 1.3511, + "step": 16 + }, + { + "epoch": 0.07157894736842105, + "grad_norm": 0.390625, + "learning_rate": 0.0001999725185109816, + "loss": 1.2886, + "step": 17 + }, + { + "epoch": 0.07578947368421053, + "grad_norm": 0.388671875, + "learning_rate": 0.00019996410631356498, + "loss": 1.3057, + "step": 18 + }, + { + "epoch": 0.08, + "grad_norm": 0.6328125, + "learning_rate": 0.00019995457277502271, + "loss": 1.7375, + "step": 19 + }, + { + "epoch": 0.08421052631578947, + "grad_norm": 0.51953125, + "learning_rate": 0.00019994391800229666, + "loss": 1.3436, + "step": 20 + }, + { + "epoch": 0.08842105263157894, + "grad_norm": 0.345703125, + "learning_rate": 0.00019993214211490608, + "loss": 1.14, + "step": 21 + }, + { + "epoch": 0.09263157894736843, + "grad_norm": 0.357421875, + "learning_rate": 0.00019991924524494627, + "loss": 1.1322, + "step": 22 + }, + { + "epoch": 0.0968421052631579, + "grad_norm": 0.431640625, + "learning_rate": 0.00019990522753708705, + "loss": 1.4512, + "step": 23 + }, + { + "epoch": 0.10105263157894737, + "grad_norm": 0.2490234375, + "learning_rate": 0.00019989008914857116, + "loss": 1.3419, + "step": 24 + }, + { + "epoch": 0.10526315789473684, + "grad_norm": 0.78515625, + "learning_rate": 0.00019987383024921255, + "loss": 1.382, + "step": 25 + }, + { + "epoch": 0.10947368421052632, + "grad_norm": 0.56640625, + "learning_rate": 0.0001998564510213944, + "loss": 1.582, + "step": 26 + }, + { + "epoch": 0.11368421052631579, + "grad_norm": 0.5703125, + "learning_rate": 0.00019983795166006715, + "loss": 1.188, + "step": 27 + }, + { + "epoch": 0.11789473684210526, + "grad_norm": 0.46875, + "learning_rate": 0.00019981833237274618, + "loss": 1.4246, + "step": 28 + }, + { + "epoch": 0.12210526315789473, + "grad_norm": 0.3984375, + "learning_rate": 0.00019979759337950968, + "loss": 1.2828, + "step": 29 + }, + { + "epoch": 0.12631578947368421, + "grad_norm": 0.55078125, + "learning_rate": 0.00019977573491299598, + "loss": 1.6199, + "step": 30 + }, + { + "epoch": 0.13052631578947368, + "grad_norm": 0.8359375, + "learning_rate": 0.00019975275721840103, + "loss": 1.2398, + "step": 31 + }, + { + "epoch": 0.13473684210526315, + "grad_norm": 0.5, + "learning_rate": 0.00019972866055347572, + "loss": 1.4295, + "step": 32 + }, + { + "epoch": 0.13894736842105262, + "grad_norm": 0.5390625, + "learning_rate": 0.0001997034451885229, + "loss": 1.2814, + "step": 33 + }, + { + "epoch": 0.1431578947368421, + "grad_norm": 0.28515625, + "learning_rate": 0.0001996771114063943, + "loss": 1.532, + "step": 34 + }, + { + "epoch": 0.14736842105263157, + "grad_norm": 0.40625, + "learning_rate": 0.00019964965950248757, + "loss": 1.3516, + "step": 35 + }, + { + "epoch": 0.15157894736842106, + "grad_norm": 0.44921875, + "learning_rate": 0.00019962108978474263, + "loss": 1.1487, + "step": 36 + }, + { + "epoch": 0.15578947368421053, + "grad_norm": 0.310546875, + "learning_rate": 0.0001995914025736386, + "loss": 1.3543, + "step": 37 + }, + { + "epoch": 0.16, + "grad_norm": 0.306640625, + "learning_rate": 0.00019956059820218982, + "loss": 1.2986, + "step": 38 + }, + { + "epoch": 0.16421052631578947, + "grad_norm": 0.625, + "learning_rate": 0.00019952867701594247, + "loss": 1.0167, + "step": 39 + }, + { + "epoch": 0.16842105263157894, + "grad_norm": 0.306640625, + "learning_rate": 0.00019949563937297045, + "loss": 1.2287, + "step": 40 + }, + { + "epoch": 0.1726315789473684, + "grad_norm": 0.34765625, + "learning_rate": 0.00019946148564387145, + "loss": 1.0063, + "step": 41 + }, + { + "epoch": 0.17684210526315788, + "grad_norm": 0.94921875, + "learning_rate": 0.00019942621621176282, + "loss": 1.2712, + "step": 42 + }, + { + "epoch": 0.18105263157894738, + "grad_norm": 0.333984375, + "learning_rate": 0.00019938983147227716, + "loss": 1.2106, + "step": 43 + }, + { + "epoch": 0.18526315789473685, + "grad_norm": 0.6328125, + "learning_rate": 0.0001993523318335581, + "loss": 1.4887, + "step": 44 + }, + { + "epoch": 0.18947368421052632, + "grad_norm": 0.62890625, + "learning_rate": 0.00019931371771625544, + "loss": 1.2747, + "step": 45 + }, + { + "epoch": 0.1936842105263158, + "grad_norm": 0.2216796875, + "learning_rate": 0.00019927398955352061, + "loss": 1.1946, + "step": 46 + }, + { + "epoch": 0.19789473684210526, + "grad_norm": 0.54296875, + "learning_rate": 0.00019923314779100184, + "loss": 1.2467, + "step": 47 + }, + { + "epoch": 0.20210526315789473, + "grad_norm": 0.25390625, + "learning_rate": 0.00019919119288683908, + "loss": 1.5067, + "step": 48 + }, + { + "epoch": 0.2063157894736842, + "grad_norm": 0.404296875, + "learning_rate": 0.00019914812531165877, + "loss": 1.4291, + "step": 49 + }, + { + "epoch": 0.21052631578947367, + "grad_norm": 0.267578125, + "learning_rate": 0.00019910394554856876, + "loss": 1.0136, + "step": 50 + }, + { + "epoch": 0.21473684210526317, + "grad_norm": 1.328125, + "learning_rate": 0.00019905865409315288, + "loss": 1.2874, + "step": 51 + }, + { + "epoch": 0.21894736842105264, + "grad_norm": 0.244140625, + "learning_rate": 0.0001990122514534651, + "loss": 0.9588, + "step": 52 + }, + { + "epoch": 0.2231578947368421, + "grad_norm": 0.376953125, + "learning_rate": 0.0001989647381500242, + "loss": 1.3036, + "step": 53 + }, + { + "epoch": 0.22736842105263158, + "grad_norm": 0.5, + "learning_rate": 0.00019891611471580764, + "loss": 1.514, + "step": 54 + }, + { + "epoch": 0.23157894736842105, + "grad_norm": 0.423828125, + "learning_rate": 0.0001988663816962458, + "loss": 1.2444, + "step": 55 + }, + { + "epoch": 0.23578947368421052, + "grad_norm": 0.30078125, + "learning_rate": 0.00019881553964921572, + "loss": 1.2261, + "step": 56 + }, + { + "epoch": 0.24, + "grad_norm": 0.34765625, + "learning_rate": 0.00019876358914503492, + "loss": 1.2489, + "step": 57 + }, + { + "epoch": 0.24421052631578946, + "grad_norm": 0.431640625, + "learning_rate": 0.00019871053076645488, + "loss": 1.2134, + "step": 58 + }, + { + "epoch": 0.24842105263157896, + "grad_norm": 0.259765625, + "learning_rate": 0.00019865636510865467, + "loss": 1.1999, + "step": 59 + }, + { + "epoch": 0.25263157894736843, + "grad_norm": 0.2421875, + "learning_rate": 0.00019860109277923418, + "loss": 1.1485, + "step": 60 + }, + { + "epoch": 0.25263157894736843, + "eval_loss": 1.2386109828948975, + "eval_runtime": 25.8708, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.933, + "step": 60 + }, + { + "epoch": 0.25684210526315787, + "grad_norm": 0.5390625, + "learning_rate": 0.00019854471439820733, + "loss": 1.2963, + "step": 61 + }, + { + "epoch": 0.26105263157894737, + "grad_norm": 0.345703125, + "learning_rate": 0.00019848723059799506, + "loss": 1.097, + "step": 62 + }, + { + "epoch": 0.26526315789473687, + "grad_norm": 0.49609375, + "learning_rate": 0.00019842864202341835, + "loss": 1.1463, + "step": 63 + }, + { + "epoch": 0.2694736842105263, + "grad_norm": 0.267578125, + "learning_rate": 0.00019836894933169088, + "loss": 1.1921, + "step": 64 + }, + { + "epoch": 0.2736842105263158, + "grad_norm": 0.349609375, + "learning_rate": 0.00019830815319241168, + "loss": 1.1835, + "step": 65 + }, + { + "epoch": 0.27789473684210525, + "grad_norm": 0.333984375, + "learning_rate": 0.0001982462542875576, + "loss": 1.1998, + "step": 66 + }, + { + "epoch": 0.28210526315789475, + "grad_norm": 0.244140625, + "learning_rate": 0.00019818325331147585, + "loss": 1.1047, + "step": 67 + }, + { + "epoch": 0.2863157894736842, + "grad_norm": 0.244140625, + "learning_rate": 0.00019811915097087587, + "loss": 1.2452, + "step": 68 + }, + { + "epoch": 0.2905263157894737, + "grad_norm": 0.259765625, + "learning_rate": 0.00019805394798482167, + "loss": 1.1633, + "step": 69 + }, + { + "epoch": 0.29473684210526313, + "grad_norm": 0.44140625, + "learning_rate": 0.00019798764508472373, + "loss": 1.0081, + "step": 70 + }, + { + "epoch": 0.29894736842105263, + "grad_norm": 0.5703125, + "learning_rate": 0.00019792024301433064, + "loss": 0.9194, + "step": 71 + }, + { + "epoch": 0.3031578947368421, + "grad_norm": 0.357421875, + "learning_rate": 0.00019785174252972092, + "loss": 0.8091, + "step": 72 + }, + { + "epoch": 0.30736842105263157, + "grad_norm": 0.33984375, + "learning_rate": 0.00019778214439929452, + "loss": 1.1248, + "step": 73 + }, + { + "epoch": 0.31157894736842107, + "grad_norm": 0.23046875, + "learning_rate": 0.0001977114494037641, + "loss": 1.1204, + "step": 74 + }, + { + "epoch": 0.3157894736842105, + "grad_norm": 0.51171875, + "learning_rate": 0.0001976396583361464, + "loss": 1.1662, + "step": 75 + }, + { + "epoch": 0.32, + "grad_norm": 0.2578125, + "learning_rate": 0.00019756677200175315, + "loss": 1.1222, + "step": 76 + }, + { + "epoch": 0.32421052631578945, + "grad_norm": 0.796875, + "learning_rate": 0.00019749279121818235, + "loss": 1.0329, + "step": 77 + }, + { + "epoch": 0.32842105263157895, + "grad_norm": 0.2490234375, + "learning_rate": 0.0001974177168153088, + "loss": 1.1199, + "step": 78 + }, + { + "epoch": 0.33263157894736844, + "grad_norm": 0.26171875, + "learning_rate": 0.00019734154963527495, + "loss": 1.4391, + "step": 79 + }, + { + "epoch": 0.3368421052631579, + "grad_norm": 0.30078125, + "learning_rate": 0.0001972642905324813, + "loss": 1.3924, + "step": 80 + }, + { + "epoch": 0.3410526315789474, + "grad_norm": 0.208984375, + "learning_rate": 0.00019718594037357713, + "loss": 1.1518, + "step": 81 + }, + { + "epoch": 0.3452631578947368, + "grad_norm": 0.953125, + "learning_rate": 0.0001971065000374504, + "loss": 1.2606, + "step": 82 + }, + { + "epoch": 0.3494736842105263, + "grad_norm": 0.306640625, + "learning_rate": 0.0001970259704152181, + "loss": 1.2293, + "step": 83 + }, + { + "epoch": 0.35368421052631577, + "grad_norm": 0.6328125, + "learning_rate": 0.0001969443524102163, + "loss": 0.9564, + "step": 84 + }, + { + "epoch": 0.35789473684210527, + "grad_norm": 0.279296875, + "learning_rate": 0.00019686164693798986, + "loss": 1.2023, + "step": 85 + }, + { + "epoch": 0.36210526315789476, + "grad_norm": 0.263671875, + "learning_rate": 0.0001967778549262822, + "loss": 0.9322, + "step": 86 + }, + { + "epoch": 0.3663157894736842, + "grad_norm": 0.51171875, + "learning_rate": 0.00019669297731502507, + "loss": 1.2589, + "step": 87 + }, + { + "epoch": 0.3705263157894737, + "grad_norm": 0.392578125, + "learning_rate": 0.00019660701505632772, + "loss": 1.3188, + "step": 88 + }, + { + "epoch": 0.37473684210526315, + "grad_norm": 0.51953125, + "learning_rate": 0.00019651996911446647, + "loss": 1.0415, + "step": 89 + }, + { + "epoch": 0.37894736842105264, + "grad_norm": 0.427734375, + "learning_rate": 0.0001964318404658737, + "loss": 1.467, + "step": 90 + }, + { + "epoch": 0.3831578947368421, + "grad_norm": 0.4453125, + "learning_rate": 0.00019634263009912705, + "loss": 1.5527, + "step": 91 + }, + { + "epoch": 0.3873684210526316, + "grad_norm": 0.30859375, + "learning_rate": 0.00019625233901493822, + "loss": 1.2043, + "step": 92 + }, + { + "epoch": 0.391578947368421, + "grad_norm": 0.279296875, + "learning_rate": 0.00019616096822614186, + "loss": 0.9677, + "step": 93 + }, + { + "epoch": 0.3957894736842105, + "grad_norm": 0.46484375, + "learning_rate": 0.000196068518757684, + "loss": 1.1942, + "step": 94 + }, + { + "epoch": 0.4, + "grad_norm": 0.435546875, + "learning_rate": 0.00019597499164661093, + "loss": 1.3481, + "step": 95 + }, + { + "epoch": 0.40421052631578946, + "grad_norm": 0.291015625, + "learning_rate": 0.00019588038794205703, + "loss": 1.2776, + "step": 96 + }, + { + "epoch": 0.40842105263157896, + "grad_norm": 0.3828125, + "learning_rate": 0.00019578470870523359, + "loss": 1.4808, + "step": 97 + }, + { + "epoch": 0.4126315789473684, + "grad_norm": 0.28125, + "learning_rate": 0.00019568795500941635, + "loss": 1.3385, + "step": 98 + }, + { + "epoch": 0.4168421052631579, + "grad_norm": 0.484375, + "learning_rate": 0.00019559012793993392, + "loss": 1.5787, + "step": 99 + }, + { + "epoch": 0.42105263157894735, + "grad_norm": 0.298828125, + "learning_rate": 0.00019549122859415538, + "loss": 1.2151, + "step": 100 + }, + { + "epoch": 0.42526315789473684, + "grad_norm": 0.65234375, + "learning_rate": 0.0001953912580814779, + "loss": 1.0332, + "step": 101 + }, + { + "epoch": 0.42947368421052634, + "grad_norm": 0.37890625, + "learning_rate": 0.00019529021752331453, + "loss": 1.2677, + "step": 102 + }, + { + "epoch": 0.4336842105263158, + "grad_norm": 0.265625, + "learning_rate": 0.0001951881080530814, + "loss": 1.0672, + "step": 103 + }, + { + "epoch": 0.4378947368421053, + "grad_norm": 0.498046875, + "learning_rate": 0.00019508493081618513, + "loss": 1.1736, + "step": 104 + }, + { + "epoch": 0.4421052631578947, + "grad_norm": 0.298828125, + "learning_rate": 0.00019498068697000994, + "loss": 1.1174, + "step": 105 + }, + { + "epoch": 0.4463157894736842, + "grad_norm": 0.169921875, + "learning_rate": 0.00019487537768390464, + "loss": 1.0712, + "step": 106 + }, + { + "epoch": 0.45052631578947366, + "grad_norm": 0.55078125, + "learning_rate": 0.0001947690041391696, + "loss": 1.4496, + "step": 107 + }, + { + "epoch": 0.45473684210526316, + "grad_norm": 0.47265625, + "learning_rate": 0.00019466156752904343, + "loss": 1.2413, + "step": 108 + }, + { + "epoch": 0.4589473684210526, + "grad_norm": 0.6015625, + "learning_rate": 0.0001945530690586896, + "loss": 1.0208, + "step": 109 + }, + { + "epoch": 0.4631578947368421, + "grad_norm": 0.35546875, + "learning_rate": 0.0001944435099451829, + "loss": 1.3997, + "step": 110 + }, + { + "epoch": 0.4673684210526316, + "grad_norm": 0.330078125, + "learning_rate": 0.00019433289141749586, + "loss": 1.2914, + "step": 111 + }, + { + "epoch": 0.47157894736842104, + "grad_norm": 0.63671875, + "learning_rate": 0.00019422121471648497, + "loss": 0.8134, + "step": 112 + }, + { + "epoch": 0.47578947368421054, + "grad_norm": 0.30859375, + "learning_rate": 0.00019410848109487658, + "loss": 1.1282, + "step": 113 + }, + { + "epoch": 0.48, + "grad_norm": 0.5625, + "learning_rate": 0.0001939946918172531, + "loss": 1.3254, + "step": 114 + }, + { + "epoch": 0.4842105263157895, + "grad_norm": 0.30859375, + "learning_rate": 0.00019387984816003867, + "loss": 1.6428, + "step": 115 + }, + { + "epoch": 0.4884210526315789, + "grad_norm": 0.33203125, + "learning_rate": 0.00019376395141148476, + "loss": 1.0688, + "step": 116 + }, + { + "epoch": 0.4926315789473684, + "grad_norm": 0.6953125, + "learning_rate": 0.00019364700287165604, + "loss": 1.0969, + "step": 117 + }, + { + "epoch": 0.4968421052631579, + "grad_norm": 1.296875, + "learning_rate": 0.00019352900385241536, + "loss": 1.5249, + "step": 118 + }, + { + "epoch": 0.5010526315789474, + "grad_norm": 0.2001953125, + "learning_rate": 0.00019340995567740938, + "loss": 1.2095, + "step": 119 + }, + { + "epoch": 0.5052631578947369, + "grad_norm": 0.53515625, + "learning_rate": 0.0001932898596820536, + "loss": 1.1249, + "step": 120 + }, + { + "epoch": 0.5052631578947369, + "eval_loss": 1.2166697978973389, + "eval_runtime": 25.8713, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.933, + "step": 120 + }, + { + "epoch": 0.5094736842105263, + "grad_norm": 0.423828125, + "learning_rate": 0.00019316871721351733, + "loss": 1.5017, + "step": 121 + }, + { + "epoch": 0.5136842105263157, + "grad_norm": 0.7421875, + "learning_rate": 0.0001930465296307087, + "loss": 1.0994, + "step": 122 + }, + { + "epoch": 0.5178947368421053, + "grad_norm": 0.21875, + "learning_rate": 0.0001929232983042592, + "loss": 1.0339, + "step": 123 + }, + { + "epoch": 0.5221052631578947, + "grad_norm": 0.6328125, + "learning_rate": 0.00019279902461650866, + "loss": 1.1317, + "step": 124 + }, + { + "epoch": 0.5263157894736842, + "grad_norm": 0.3984375, + "learning_rate": 0.00019267370996148934, + "loss": 0.9421, + "step": 125 + }, + { + "epoch": 0.5305263157894737, + "grad_norm": 0.396484375, + "learning_rate": 0.00019254735574491058, + "loss": 1.3029, + "step": 126 + }, + { + "epoch": 0.5347368421052632, + "grad_norm": 0.8359375, + "learning_rate": 0.00019241996338414296, + "loss": 1.3548, + "step": 127 + }, + { + "epoch": 0.5389473684210526, + "grad_norm": 0.84765625, + "learning_rate": 0.00019229153430820232, + "loss": 1.0865, + "step": 128 + }, + { + "epoch": 0.5431578947368421, + "grad_norm": 0.453125, + "learning_rate": 0.00019216206995773373, + "loss": 0.8151, + "step": 129 + }, + { + "epoch": 0.5473684210526316, + "grad_norm": 0.8046875, + "learning_rate": 0.0001920315717849956, + "loss": 1.3576, + "step": 130 + }, + { + "epoch": 0.5515789473684211, + "grad_norm": 0.82421875, + "learning_rate": 0.00019190004125384293, + "loss": 1.2439, + "step": 131 + }, + { + "epoch": 0.5557894736842105, + "grad_norm": 0.56640625, + "learning_rate": 0.0001917674798397113, + "loss": 0.9743, + "step": 132 + }, + { + "epoch": 0.56, + "grad_norm": 1.1015625, + "learning_rate": 0.0001916338890296001, + "loss": 1.3315, + "step": 133 + }, + { + "epoch": 0.5642105263157895, + "grad_norm": 0.333984375, + "learning_rate": 0.00019149927032205587, + "loss": 1.3268, + "step": 134 + }, + { + "epoch": 0.5684210526315789, + "grad_norm": 0.388671875, + "learning_rate": 0.00019136362522715562, + "loss": 1.4751, + "step": 135 + }, + { + "epoch": 0.5726315789473684, + "grad_norm": 0.2734375, + "learning_rate": 0.00019122695526648968, + "loss": 1.1219, + "step": 136 + }, + { + "epoch": 0.5768421052631579, + "grad_norm": 1.0703125, + "learning_rate": 0.00019108926197314486, + "loss": 1.4059, + "step": 137 + }, + { + "epoch": 0.5810526315789474, + "grad_norm": 0.33203125, + "learning_rate": 0.00019095054689168705, + "loss": 1.3282, + "step": 138 + }, + { + "epoch": 0.5852631578947368, + "grad_norm": 0.6171875, + "learning_rate": 0.00019081081157814404, + "loss": 1.6334, + "step": 139 + }, + { + "epoch": 0.5894736842105263, + "grad_norm": 0.359375, + "learning_rate": 0.00019067005759998797, + "loss": 1.159, + "step": 140 + }, + { + "epoch": 0.5936842105263158, + "grad_norm": 0.470703125, + "learning_rate": 0.00019052828653611786, + "loss": 1.3261, + "step": 141 + }, + { + "epoch": 0.5978947368421053, + "grad_norm": 0.26171875, + "learning_rate": 0.0001903854999768417, + "loss": 1.2041, + "step": 142 + }, + { + "epoch": 0.6021052631578947, + "grad_norm": 0.27734375, + "learning_rate": 0.00019024169952385885, + "loss": 1.1998, + "step": 143 + }, + { + "epoch": 0.6063157894736843, + "grad_norm": 0.56640625, + "learning_rate": 0.0001900968867902419, + "loss": 1.6427, + "step": 144 + }, + { + "epoch": 0.6105263157894737, + "grad_norm": 0.419921875, + "learning_rate": 0.00018995106340041873, + "loss": 1.1968, + "step": 145 + }, + { + "epoch": 0.6147368421052631, + "grad_norm": 0.671875, + "learning_rate": 0.00018980423099015402, + "loss": 0.8641, + "step": 146 + }, + { + "epoch": 0.6189473684210526, + "grad_norm": 0.333984375, + "learning_rate": 0.00018965639120653121, + "loss": 1.4518, + "step": 147 + }, + { + "epoch": 0.6231578947368421, + "grad_norm": 0.423828125, + "learning_rate": 0.00018950754570793384, + "loss": 1.3077, + "step": 148 + }, + { + "epoch": 0.6273684210526316, + "grad_norm": 0.85546875, + "learning_rate": 0.00018935769616402703, + "loss": 1.2077, + "step": 149 + }, + { + "epoch": 0.631578947368421, + "grad_norm": 0.224609375, + "learning_rate": 0.00018920684425573865, + "loss": 1.0868, + "step": 150 + }, + { + "epoch": 0.6357894736842106, + "grad_norm": 0.25390625, + "learning_rate": 0.00018905499167524062, + "loss": 1.2018, + "step": 151 + }, + { + "epoch": 0.64, + "grad_norm": 0.498046875, + "learning_rate": 0.00018890214012592975, + "loss": 1.1675, + "step": 152 + }, + { + "epoch": 0.6442105263157895, + "grad_norm": 0.2451171875, + "learning_rate": 0.00018874829132240874, + "loss": 1.2348, + "step": 153 + }, + { + "epoch": 0.6484210526315789, + "grad_norm": 0.33984375, + "learning_rate": 0.000188593446990467, + "loss": 1.3202, + "step": 154 + }, + { + "epoch": 0.6526315789473685, + "grad_norm": 1.0078125, + "learning_rate": 0.0001884376088670611, + "loss": 1.0171, + "step": 155 + }, + { + "epoch": 0.6568421052631579, + "grad_norm": 0.318359375, + "learning_rate": 0.00018828077870029552, + "loss": 1.219, + "step": 156 + }, + { + "epoch": 0.6610526315789473, + "grad_norm": 0.369140625, + "learning_rate": 0.00018812295824940285, + "loss": 0.8956, + "step": 157 + }, + { + "epoch": 0.6652631578947369, + "grad_norm": 0.328125, + "learning_rate": 0.00018796414928472417, + "loss": 0.9322, + "step": 158 + }, + { + "epoch": 0.6694736842105263, + "grad_norm": 0.2890625, + "learning_rate": 0.0001878043535876891, + "loss": 1.2062, + "step": 159 + }, + { + "epoch": 0.6736842105263158, + "grad_norm": 0.41015625, + "learning_rate": 0.0001876435729507959, + "loss": 1.3113, + "step": 160 + }, + { + "epoch": 0.6778947368421052, + "grad_norm": 0.53515625, + "learning_rate": 0.00018748180917759136, + "loss": 1.1539, + "step": 161 + }, + { + "epoch": 0.6821052631578948, + "grad_norm": 0.279296875, + "learning_rate": 0.0001873190640826505, + "loss": 1.3287, + "step": 162 + }, + { + "epoch": 0.6863157894736842, + "grad_norm": 0.298828125, + "learning_rate": 0.00018715533949155625, + "loss": 1.0616, + "step": 163 + }, + { + "epoch": 0.6905263157894737, + "grad_norm": 0.388671875, + "learning_rate": 0.00018699063724087904, + "loss": 1.0171, + "step": 164 + }, + { + "epoch": 0.6947368421052632, + "grad_norm": 0.32421875, + "learning_rate": 0.00018682495917815608, + "loss": 1.1113, + "step": 165 + }, + { + "epoch": 0.6989473684210527, + "grad_norm": 0.46875, + "learning_rate": 0.00018665830716187065, + "loss": 0.8351, + "step": 166 + }, + { + "epoch": 0.7031578947368421, + "grad_norm": 0.6015625, + "learning_rate": 0.0001864906830614314, + "loss": 1.2181, + "step": 167 + }, + { + "epoch": 0.7073684210526315, + "grad_norm": 0.71875, + "learning_rate": 0.0001863220887571512, + "loss": 1.1843, + "step": 168 + }, + { + "epoch": 0.7115789473684211, + "grad_norm": 0.51953125, + "learning_rate": 0.00018615252614022617, + "loss": 1.3809, + "step": 169 + }, + { + "epoch": 0.7157894736842105, + "grad_norm": 0.34375, + "learning_rate": 0.0001859819971127143, + "loss": 1.4288, + "step": 170 + }, + { + "epoch": 0.72, + "grad_norm": 0.41796875, + "learning_rate": 0.00018581050358751445, + "loss": 1.1449, + "step": 171 + }, + { + "epoch": 0.7242105263157895, + "grad_norm": 0.578125, + "learning_rate": 0.00018563804748834438, + "loss": 1.0703, + "step": 172 + }, + { + "epoch": 0.728421052631579, + "grad_norm": 0.43359375, + "learning_rate": 0.00018546463074971977, + "loss": 1.2093, + "step": 173 + }, + { + "epoch": 0.7326315789473684, + "grad_norm": 0.26171875, + "learning_rate": 0.000185290255316932, + "loss": 0.7346, + "step": 174 + }, + { + "epoch": 0.7368421052631579, + "grad_norm": 0.369140625, + "learning_rate": 0.00018511492314602656, + "loss": 1.2782, + "step": 175 + }, + { + "epoch": 0.7410526315789474, + "grad_norm": 0.56640625, + "learning_rate": 0.00018493863620378122, + "loss": 1.0844, + "step": 176 + }, + { + "epoch": 0.7452631578947368, + "grad_norm": 0.2470703125, + "learning_rate": 0.00018476139646768372, + "loss": 1.1771, + "step": 177 + }, + { + "epoch": 0.7494736842105263, + "grad_norm": 0.228515625, + "learning_rate": 0.00018458320592590975, + "loss": 1.2, + "step": 178 + }, + { + "epoch": 0.7536842105263157, + "grad_norm": 0.310546875, + "learning_rate": 0.00018440406657730058, + "loss": 1.2248, + "step": 179 + }, + { + "epoch": 0.7578947368421053, + "grad_norm": 0.263671875, + "learning_rate": 0.00018422398043134067, + "loss": 1.3675, + "step": 180 + }, + { + "epoch": 0.7578947368421053, + "eval_loss": 1.2129631042480469, + "eval_runtime": 25.9253, + "eval_samples_per_second": 3.857, + "eval_steps_per_second": 1.929, + "step": 180 + }, + { + "epoch": 0.7621052631578947, + "grad_norm": 0.2412109375, + "learning_rate": 0.0001840429495081352, + "loss": 1.2456, + "step": 181 + }, + { + "epoch": 0.7663157894736842, + "grad_norm": 0.25390625, + "learning_rate": 0.00018386097583838714, + "loss": 1.2617, + "step": 182 + }, + { + "epoch": 0.7705263157894737, + "grad_norm": 0.283203125, + "learning_rate": 0.00018367806146337483, + "loss": 1.2708, + "step": 183 + }, + { + "epoch": 0.7747368421052632, + "grad_norm": 0.318359375, + "learning_rate": 0.00018349420843492888, + "loss": 1.3771, + "step": 184 + }, + { + "epoch": 0.7789473684210526, + "grad_norm": 0.73046875, + "learning_rate": 0.00018330941881540915, + "loss": 1.3123, + "step": 185 + }, + { + "epoch": 0.783157894736842, + "grad_norm": 0.46875, + "learning_rate": 0.00018312369467768166, + "loss": 1.1952, + "step": 186 + }, + { + "epoch": 0.7873684210526316, + "grad_norm": 0.7578125, + "learning_rate": 0.00018293703810509537, + "loss": 1.426, + "step": 187 + }, + { + "epoch": 0.791578947368421, + "grad_norm": 0.33203125, + "learning_rate": 0.0001827494511914587, + "loss": 1.3205, + "step": 188 + }, + { + "epoch": 0.7957894736842105, + "grad_norm": 0.365234375, + "learning_rate": 0.00018256093604101617, + "loss": 1.2221, + "step": 189 + }, + { + "epoch": 0.8, + "grad_norm": 0.36328125, + "learning_rate": 0.0001823714947684247, + "loss": 1.4212, + "step": 190 + }, + { + "epoch": 0.8042105263157895, + "grad_norm": 0.349609375, + "learning_rate": 0.00018218112949872995, + "loss": 1.2183, + "step": 191 + }, + { + "epoch": 0.8084210526315789, + "grad_norm": 0.392578125, + "learning_rate": 0.00018198984236734246, + "loss": 0.9922, + "step": 192 + }, + { + "epoch": 0.8126315789473684, + "grad_norm": 0.373046875, + "learning_rate": 0.0001817976355200137, + "loss": 1.334, + "step": 193 + }, + { + "epoch": 0.8168421052631579, + "grad_norm": 0.30859375, + "learning_rate": 0.000181604511112812, + "loss": 1.1052, + "step": 194 + }, + { + "epoch": 0.8210526315789474, + "grad_norm": 0.349609375, + "learning_rate": 0.0001814104713120983, + "loss": 0.8112, + "step": 195 + }, + { + "epoch": 0.8252631578947368, + "grad_norm": 0.35546875, + "learning_rate": 0.000181215518294502, + "loss": 1.2507, + "step": 196 + }, + { + "epoch": 0.8294736842105264, + "grad_norm": 0.671875, + "learning_rate": 0.0001810196542468964, + "loss": 1.1228, + "step": 197 + }, + { + "epoch": 0.8336842105263158, + "grad_norm": 0.41015625, + "learning_rate": 0.00018082288136637422, + "loss": 1.3562, + "step": 198 + }, + { + "epoch": 0.8378947368421052, + "grad_norm": 0.302734375, + "learning_rate": 0.000180625201860223, + "loss": 0.7766, + "step": 199 + }, + { + "epoch": 0.8421052631578947, + "grad_norm": 0.322265625, + "learning_rate": 0.00018042661794590023, + "loss": 1.2885, + "step": 200 + }, + { + "epoch": 0.8463157894736842, + "grad_norm": 0.609375, + "learning_rate": 0.0001802271318510086, + "loss": 1.271, + "step": 201 + }, + { + "epoch": 0.8505263157894737, + "grad_norm": 0.341796875, + "learning_rate": 0.00018002674581327094, + "loss": 1.3777, + "step": 202 + }, + { + "epoch": 0.8547368421052631, + "grad_norm": 0.3828125, + "learning_rate": 0.00017982546208050515, + "loss": 1.32, + "step": 203 + }, + { + "epoch": 0.8589473684210527, + "grad_norm": 0.392578125, + "learning_rate": 0.00017962328291059888, + "loss": 1.1265, + "step": 204 + }, + { + "epoch": 0.8631578947368421, + "grad_norm": 0.73046875, + "learning_rate": 0.00017942021057148436, + "loss": 1.523, + "step": 205 + }, + { + "epoch": 0.8673684210526316, + "grad_norm": 1.046875, + "learning_rate": 0.00017921624734111292, + "loss": 1.0251, + "step": 206 + }, + { + "epoch": 0.871578947368421, + "grad_norm": 0.490234375, + "learning_rate": 0.0001790113955074294, + "loss": 1.2665, + "step": 207 + }, + { + "epoch": 0.8757894736842106, + "grad_norm": 0.6875, + "learning_rate": 0.0001788056573683464, + "loss": 1.2244, + "step": 208 + }, + { + "epoch": 0.88, + "grad_norm": 0.2177734375, + "learning_rate": 0.0001785990352317187, + "loss": 1.1621, + "step": 209 + }, + { + "epoch": 0.8842105263157894, + "grad_norm": 0.4921875, + "learning_rate": 0.00017839153141531718, + "loss": 1.3182, + "step": 210 + }, + { + "epoch": 0.888421052631579, + "grad_norm": 0.388671875, + "learning_rate": 0.000178183148246803, + "loss": 1.4508, + "step": 211 + }, + { + "epoch": 0.8926315789473684, + "grad_norm": 0.388671875, + "learning_rate": 0.00017797388806370132, + "loss": 1.1155, + "step": 212 + }, + { + "epoch": 0.8968421052631579, + "grad_norm": 0.416015625, + "learning_rate": 0.00017776375321337521, + "loss": 0.9191, + "step": 213 + }, + { + "epoch": 0.9010526315789473, + "grad_norm": 0.2431640625, + "learning_rate": 0.00017755274605299923, + "loss": 1.1577, + "step": 214 + }, + { + "epoch": 0.9052631578947369, + "grad_norm": 0.412109375, + "learning_rate": 0.00017734086894953305, + "loss": 1.372, + "step": 215 + }, + { + "epoch": 0.9094736842105263, + "grad_norm": 0.353515625, + "learning_rate": 0.00017712812427969485, + "loss": 1.1932, + "step": 216 + }, + { + "epoch": 0.9136842105263158, + "grad_norm": 0.291015625, + "learning_rate": 0.0001769145144299347, + "loss": 1.3167, + "step": 217 + }, + { + "epoch": 0.9178947368421052, + "grad_norm": 0.255859375, + "learning_rate": 0.00017670004179640774, + "loss": 1.2357, + "step": 218 + }, + { + "epoch": 0.9221052631578948, + "grad_norm": 0.5703125, + "learning_rate": 0.00017648470878494737, + "loss": 0.9556, + "step": 219 + }, + { + "epoch": 0.9263157894736842, + "grad_norm": 0.36328125, + "learning_rate": 0.0001762685178110382, + "loss": 1.2047, + "step": 220 + }, + { + "epoch": 0.9305263157894736, + "grad_norm": 0.271484375, + "learning_rate": 0.00017605147129978903, + "loss": 1.1935, + "step": 221 + }, + { + "epoch": 0.9347368421052632, + "grad_norm": 0.400390625, + "learning_rate": 0.0001758335716859055, + "loss": 1.3426, + "step": 222 + }, + { + "epoch": 0.9389473684210526, + "grad_norm": 0.73046875, + "learning_rate": 0.000175614821413663, + "loss": 1.5181, + "step": 223 + }, + { + "epoch": 0.9431578947368421, + "grad_norm": 0.3203125, + "learning_rate": 0.00017539522293687898, + "loss": 1.0962, + "step": 224 + }, + { + "epoch": 0.9473684210526315, + "grad_norm": 0.232421875, + "learning_rate": 0.00017517477871888568, + "loss": 1.3922, + "step": 225 + }, + { + "epoch": 0.9515789473684211, + "grad_norm": 0.5546875, + "learning_rate": 0.00017495349123250242, + "loss": 1.2535, + "step": 226 + }, + { + "epoch": 0.9557894736842105, + "grad_norm": 0.296875, + "learning_rate": 0.00017473136296000772, + "loss": 1.1987, + "step": 227 + }, + { + "epoch": 0.96, + "grad_norm": 0.77734375, + "learning_rate": 0.00017450839639311162, + "loss": 1.4301, + "step": 228 + }, + { + "epoch": 0.9642105263157895, + "grad_norm": 0.35546875, + "learning_rate": 0.00017428459403292767, + "loss": 1.3164, + "step": 229 + }, + { + "epoch": 0.968421052631579, + "grad_norm": 0.419921875, + "learning_rate": 0.00017405995838994494, + "loss": 1.1554, + "step": 230 + }, + { + "epoch": 0.9726315789473684, + "grad_norm": 0.3125, + "learning_rate": 0.00017383449198399964, + "loss": 1.3036, + "step": 231 + }, + { + "epoch": 0.9768421052631578, + "grad_norm": 0.267578125, + "learning_rate": 0.00017360819734424715, + "loss": 1.2797, + "step": 232 + }, + { + "epoch": 0.9810526315789474, + "grad_norm": 0.412109375, + "learning_rate": 0.00017338107700913354, + "loss": 1.1202, + "step": 233 + }, + { + "epoch": 0.9852631578947368, + "grad_norm": 0.365234375, + "learning_rate": 0.0001731531335263669, + "loss": 0.8187, + "step": 234 + }, + { + "epoch": 0.9894736842105263, + "grad_norm": 0.5546875, + "learning_rate": 0.0001729243694528891, + "loss": 1.1702, + "step": 235 + }, + { + "epoch": 0.9936842105263158, + "grad_norm": 0.5546875, + "learning_rate": 0.00017269478735484683, + "loss": 1.1406, + "step": 236 + }, + { + "epoch": 0.9978947368421053, + "grad_norm": 0.3359375, + "learning_rate": 0.00017246438980756293, + "loss": 0.9633, + "step": 237 + } + ], + "logging_steps": 1, + "max_steps": 948, + "num_input_tokens_seen": 0, + "num_train_epochs": 4, + "save_steps": 237, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3952216796823552.0, + "train_batch_size": 2, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-237/training_args.bin b/checkpoint-237/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..bd7f9976bbc503fbce44b538eec2dade4024c02a --- /dev/null +++ b/checkpoint-237/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee017c711219d88132625f3e50d1a67d1c6236a55d3f5a2ad9f499310d968cef +size 5944 diff --git a/checkpoint-474/README.md b/checkpoint-474/README.md new file mode 100644 index 0000000000000000000000000000000000000000..e1ccd431539a8f1507d8755a9c3ba5e5b2897978 --- /dev/null +++ b/checkpoint-474/README.md @@ -0,0 +1,202 @@ +--- +library_name: peft +base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **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] +### Framework versions + +- PEFT 0.11.1 \ No newline at end of file diff --git a/checkpoint-474/adapter_config.json b/checkpoint-474/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..cfa45a23cb958fe6ad106d4adf7f6169698c1fc8 --- /dev/null +++ b/checkpoint-474/adapter_config.json @@ -0,0 +1,34 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 16, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "v_proj", + "k_proj", + "q_proj", + "gate_proj", + "up_proj", + "down_proj", + "o_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-474/adapter_model.safetensors b/checkpoint-474/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..78657aca9df2c761dbd97409172b78d212bb3e41 --- /dev/null +++ b/checkpoint-474/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:69f694edaccdd42023f0a1cee653f7f3e9fd1123a26113cdf9fd49fe00ce746e +size 50503848 diff --git a/checkpoint-474/optimizer.pt b/checkpoint-474/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..215fd3847a006d2fb09d8d0fc5847e823d5578ef --- /dev/null +++ b/checkpoint-474/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1a1e2576fd196bd384fbd66e8d7d237c7f8d9aef44cbdeac73b0d7051b70f35 +size 202035770 diff --git a/checkpoint-474/rng_state.pth b/checkpoint-474/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..f1ff410f59acd82f01e2c5757cbd889548dbbf75 --- /dev/null +++ b/checkpoint-474/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa8ea66452a468b349a3bdfd131d03d7a70b38ad4ea6f8a3c4d500229279eb91 +size 14244 diff --git a/checkpoint-474/scheduler.pt b/checkpoint-474/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..8ea95088a66431b5032d206eaeac83d4a15ee9c0 --- /dev/null +++ b/checkpoint-474/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42714fb01454b0350f3ff250f6bc8bc2da6df691649aefca623325cce43ab6b3 +size 1064 diff --git a/checkpoint-474/special_tokens_map.json b/checkpoint-474/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1 --- /dev/null +++ b/checkpoint-474/special_tokens_map.json @@ -0,0 +1,24 @@ +{ + "bos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "", + "unk_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-474/tokenizer.model b/checkpoint-474/tokenizer.model new file mode 100644 index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899 --- /dev/null +++ b/checkpoint-474/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 +size 499723 diff --git a/checkpoint-474/tokenizer_config.json b/checkpoint-474/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..0773857a13ba5a27453a0b462624fe76e8e82a86 --- /dev/null +++ b/checkpoint-474/tokenizer_config.json @@ -0,0 +1,44 @@ +{ + "add_bos_token": true, + "add_eos_token": false, + "add_prefix_space": true, + "added_tokens_decoder": { + "0": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "1": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "2": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "", + "clean_up_tokenization_spaces": false, + "eos_token": "", + "legacy": false, + "model_max_length": 1000000000000000019884624838656, + "pad_token": "", + "padding_side": "right", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "LlamaTokenizer", + "unk_token": "", + "use_default_system_prompt": false, + "use_fast": true +} diff --git a/checkpoint-474/trainer_state.json b/checkpoint-474/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..18f404990eedefb0febbd0f171b6aad733ecc1d5 --- /dev/null +++ b/checkpoint-474/trainer_state.json @@ -0,0 +1,3415 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.9957894736842106, + "eval_steps": 60, + "global_step": 474, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.004210526315789474, + "grad_norm": 0.310546875, + "learning_rate": 2e-05, + "loss": 1.5605, + "step": 1 + }, + { + "epoch": 0.004210526315789474, + "eval_loss": 1.5264556407928467, + "eval_runtime": 26.3082, + "eval_samples_per_second": 3.801, + "eval_steps_per_second": 1.901, + "step": 1 + }, + { + "epoch": 0.008421052631578947, + "grad_norm": 0.306640625, + "learning_rate": 4e-05, + "loss": 1.4702, + "step": 2 + }, + { + "epoch": 0.01263157894736842, + "grad_norm": 0.2578125, + "learning_rate": 6e-05, + "loss": 1.4576, + "step": 3 + }, + { + "epoch": 0.016842105263157894, + "grad_norm": 0.65234375, + "learning_rate": 8e-05, + "loss": 1.5868, + "step": 4 + }, + { + "epoch": 0.021052631578947368, + "grad_norm": 0.56640625, + "learning_rate": 0.0001, + "loss": 1.5916, + "step": 5 + }, + { + "epoch": 0.02526315789473684, + "grad_norm": 0.5703125, + "learning_rate": 0.00012, + "loss": 1.7851, + "step": 6 + }, + { + "epoch": 0.029473684210526315, + "grad_norm": 0.3359375, + "learning_rate": 0.00014, + "loss": 1.5319, + "step": 7 + }, + { + "epoch": 0.03368421052631579, + "grad_norm": 0.39453125, + "learning_rate": 0.00016, + "loss": 1.8462, + "step": 8 + }, + { + "epoch": 0.037894736842105266, + "grad_norm": 0.484375, + "learning_rate": 0.00018, + "loss": 1.4219, + "step": 9 + }, + { + "epoch": 0.042105263157894736, + "grad_norm": 0.30859375, + "learning_rate": 0.0002, + "loss": 1.6555, + "step": 10 + }, + { + "epoch": 0.04631578947368421, + "grad_norm": 0.267578125, + "learning_rate": 0.0001999994391281196, + "loss": 1.1527, + "step": 11 + }, + { + "epoch": 0.05052631578947368, + "grad_norm": 0.298828125, + "learning_rate": 0.00019999775651876987, + "loss": 1.4265, + "step": 12 + }, + { + "epoch": 0.05473684210526316, + "grad_norm": 0.376953125, + "learning_rate": 0.00019999495219082545, + "loss": 1.5505, + "step": 13 + }, + { + "epoch": 0.05894736842105263, + "grad_norm": 0.380859375, + "learning_rate": 0.00019999102617574365, + "loss": 1.0313, + "step": 14 + }, + { + "epoch": 0.06315789473684211, + "grad_norm": 0.494140625, + "learning_rate": 0.0001999859785175643, + "loss": 1.439, + "step": 15 + }, + { + "epoch": 0.06736842105263158, + "grad_norm": 0.40234375, + "learning_rate": 0.00019997980927290927, + "loss": 1.3511, + "step": 16 + }, + { + "epoch": 0.07157894736842105, + "grad_norm": 0.390625, + "learning_rate": 0.0001999725185109816, + "loss": 1.2886, + "step": 17 + }, + { + "epoch": 0.07578947368421053, + "grad_norm": 0.388671875, + "learning_rate": 0.00019996410631356498, + "loss": 1.3057, + "step": 18 + }, + { + "epoch": 0.08, + "grad_norm": 0.6328125, + "learning_rate": 0.00019995457277502271, + "loss": 1.7375, + "step": 19 + }, + { + "epoch": 0.08421052631578947, + "grad_norm": 0.51953125, + "learning_rate": 0.00019994391800229666, + "loss": 1.3436, + "step": 20 + }, + { + "epoch": 0.08842105263157894, + "grad_norm": 0.345703125, + "learning_rate": 0.00019993214211490608, + "loss": 1.14, + "step": 21 + }, + { + "epoch": 0.09263157894736843, + "grad_norm": 0.357421875, + "learning_rate": 0.00019991924524494627, + "loss": 1.1322, + "step": 22 + }, + { + "epoch": 0.0968421052631579, + "grad_norm": 0.431640625, + "learning_rate": 0.00019990522753708705, + "loss": 1.4512, + "step": 23 + }, + { + "epoch": 0.10105263157894737, + "grad_norm": 0.2490234375, + "learning_rate": 0.00019989008914857116, + "loss": 1.3419, + "step": 24 + }, + { + "epoch": 0.10526315789473684, + "grad_norm": 0.78515625, + "learning_rate": 0.00019987383024921255, + "loss": 1.382, + "step": 25 + }, + { + "epoch": 0.10947368421052632, + "grad_norm": 0.56640625, + "learning_rate": 0.0001998564510213944, + "loss": 1.582, + "step": 26 + }, + { + "epoch": 0.11368421052631579, + "grad_norm": 0.5703125, + "learning_rate": 0.00019983795166006715, + "loss": 1.188, + "step": 27 + }, + { + "epoch": 0.11789473684210526, + "grad_norm": 0.46875, + "learning_rate": 0.00019981833237274618, + "loss": 1.4246, + "step": 28 + }, + { + "epoch": 0.12210526315789473, + "grad_norm": 0.3984375, + "learning_rate": 0.00019979759337950968, + "loss": 1.2828, + "step": 29 + }, + { + "epoch": 0.12631578947368421, + "grad_norm": 0.55078125, + "learning_rate": 0.00019977573491299598, + "loss": 1.6199, + "step": 30 + }, + { + "epoch": 0.13052631578947368, + "grad_norm": 0.8359375, + "learning_rate": 0.00019975275721840103, + "loss": 1.2398, + "step": 31 + }, + { + "epoch": 0.13473684210526315, + "grad_norm": 0.5, + "learning_rate": 0.00019972866055347572, + "loss": 1.4295, + "step": 32 + }, + { + "epoch": 0.13894736842105262, + "grad_norm": 0.5390625, + "learning_rate": 0.0001997034451885229, + "loss": 1.2814, + "step": 33 + }, + { + "epoch": 0.1431578947368421, + "grad_norm": 0.28515625, + "learning_rate": 0.0001996771114063943, + "loss": 1.532, + "step": 34 + }, + { + "epoch": 0.14736842105263157, + "grad_norm": 0.40625, + "learning_rate": 0.00019964965950248757, + "loss": 1.3516, + "step": 35 + }, + { + "epoch": 0.15157894736842106, + "grad_norm": 0.44921875, + "learning_rate": 0.00019962108978474263, + "loss": 1.1487, + "step": 36 + }, + { + "epoch": 0.15578947368421053, + "grad_norm": 0.310546875, + "learning_rate": 0.0001995914025736386, + "loss": 1.3543, + "step": 37 + }, + { + "epoch": 0.16, + "grad_norm": 0.306640625, + "learning_rate": 0.00019956059820218982, + "loss": 1.2986, + "step": 38 + }, + { + "epoch": 0.16421052631578947, + "grad_norm": 0.625, + "learning_rate": 0.00019952867701594247, + "loss": 1.0167, + "step": 39 + }, + { + "epoch": 0.16842105263157894, + "grad_norm": 0.306640625, + "learning_rate": 0.00019949563937297045, + "loss": 1.2287, + "step": 40 + }, + { + "epoch": 0.1726315789473684, + "grad_norm": 0.34765625, + "learning_rate": 0.00019946148564387145, + "loss": 1.0063, + "step": 41 + }, + { + "epoch": 0.17684210526315788, + "grad_norm": 0.94921875, + "learning_rate": 0.00019942621621176282, + "loss": 1.2712, + "step": 42 + }, + { + "epoch": 0.18105263157894738, + "grad_norm": 0.333984375, + "learning_rate": 0.00019938983147227716, + "loss": 1.2106, + "step": 43 + }, + { + "epoch": 0.18526315789473685, + "grad_norm": 0.6328125, + "learning_rate": 0.0001993523318335581, + "loss": 1.4887, + "step": 44 + }, + { + "epoch": 0.18947368421052632, + "grad_norm": 0.62890625, + "learning_rate": 0.00019931371771625544, + "loss": 1.2747, + "step": 45 + }, + { + "epoch": 0.1936842105263158, + "grad_norm": 0.2216796875, + "learning_rate": 0.00019927398955352061, + "loss": 1.1946, + "step": 46 + }, + { + "epoch": 0.19789473684210526, + "grad_norm": 0.54296875, + "learning_rate": 0.00019923314779100184, + "loss": 1.2467, + "step": 47 + }, + { + "epoch": 0.20210526315789473, + "grad_norm": 0.25390625, + "learning_rate": 0.00019919119288683908, + "loss": 1.5067, + "step": 48 + }, + { + "epoch": 0.2063157894736842, + "grad_norm": 0.404296875, + "learning_rate": 0.00019914812531165877, + "loss": 1.4291, + "step": 49 + }, + { + "epoch": 0.21052631578947367, + "grad_norm": 0.267578125, + "learning_rate": 0.00019910394554856876, + "loss": 1.0136, + "step": 50 + }, + { + "epoch": 0.21473684210526317, + "grad_norm": 1.328125, + "learning_rate": 0.00019905865409315288, + "loss": 1.2874, + "step": 51 + }, + { + "epoch": 0.21894736842105264, + "grad_norm": 0.244140625, + "learning_rate": 0.0001990122514534651, + "loss": 0.9588, + "step": 52 + }, + { + "epoch": 0.2231578947368421, + "grad_norm": 0.376953125, + "learning_rate": 0.0001989647381500242, + "loss": 1.3036, + "step": 53 + }, + { + "epoch": 0.22736842105263158, + "grad_norm": 0.5, + "learning_rate": 0.00019891611471580764, + "loss": 1.514, + "step": 54 + }, + { + "epoch": 0.23157894736842105, + "grad_norm": 0.423828125, + "learning_rate": 0.0001988663816962458, + "loss": 1.2444, + "step": 55 + }, + { + "epoch": 0.23578947368421052, + "grad_norm": 0.30078125, + "learning_rate": 0.00019881553964921572, + "loss": 1.2261, + "step": 56 + }, + { + "epoch": 0.24, + "grad_norm": 0.34765625, + "learning_rate": 0.00019876358914503492, + "loss": 1.2489, + "step": 57 + }, + { + "epoch": 0.24421052631578946, + "grad_norm": 0.431640625, + "learning_rate": 0.00019871053076645488, + "loss": 1.2134, + "step": 58 + }, + { + "epoch": 0.24842105263157896, + "grad_norm": 0.259765625, + "learning_rate": 0.00019865636510865467, + "loss": 1.1999, + "step": 59 + }, + { + "epoch": 0.25263157894736843, + "grad_norm": 0.2421875, + "learning_rate": 0.00019860109277923418, + "loss": 1.1485, + "step": 60 + }, + { + "epoch": 0.25263157894736843, + "eval_loss": 1.2386109828948975, + "eval_runtime": 25.8708, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.933, + "step": 60 + }, + { + "epoch": 0.25684210526315787, + "grad_norm": 0.5390625, + "learning_rate": 0.00019854471439820733, + "loss": 1.2963, + "step": 61 + }, + { + "epoch": 0.26105263157894737, + "grad_norm": 0.345703125, + "learning_rate": 0.00019848723059799506, + "loss": 1.097, + "step": 62 + }, + { + "epoch": 0.26526315789473687, + "grad_norm": 0.49609375, + "learning_rate": 0.00019842864202341835, + "loss": 1.1463, + "step": 63 + }, + { + "epoch": 0.2694736842105263, + "grad_norm": 0.267578125, + "learning_rate": 0.00019836894933169088, + "loss": 1.1921, + "step": 64 + }, + { + "epoch": 0.2736842105263158, + "grad_norm": 0.349609375, + "learning_rate": 0.00019830815319241168, + "loss": 1.1835, + "step": 65 + }, + { + "epoch": 0.27789473684210525, + "grad_norm": 0.333984375, + "learning_rate": 0.0001982462542875576, + "loss": 1.1998, + "step": 66 + }, + { + "epoch": 0.28210526315789475, + "grad_norm": 0.244140625, + "learning_rate": 0.00019818325331147585, + "loss": 1.1047, + "step": 67 + }, + { + "epoch": 0.2863157894736842, + "grad_norm": 0.244140625, + "learning_rate": 0.00019811915097087587, + "loss": 1.2452, + "step": 68 + }, + { + "epoch": 0.2905263157894737, + "grad_norm": 0.259765625, + "learning_rate": 0.00019805394798482167, + "loss": 1.1633, + "step": 69 + }, + { + "epoch": 0.29473684210526313, + "grad_norm": 0.44140625, + "learning_rate": 0.00019798764508472373, + "loss": 1.0081, + "step": 70 + }, + { + "epoch": 0.29894736842105263, + "grad_norm": 0.5703125, + "learning_rate": 0.00019792024301433064, + "loss": 0.9194, + "step": 71 + }, + { + "epoch": 0.3031578947368421, + "grad_norm": 0.357421875, + "learning_rate": 0.00019785174252972092, + "loss": 0.8091, + "step": 72 + }, + { + "epoch": 0.30736842105263157, + "grad_norm": 0.33984375, + "learning_rate": 0.00019778214439929452, + "loss": 1.1248, + "step": 73 + }, + { + "epoch": 0.31157894736842107, + "grad_norm": 0.23046875, + "learning_rate": 0.0001977114494037641, + "loss": 1.1204, + "step": 74 + }, + { + "epoch": 0.3157894736842105, + "grad_norm": 0.51171875, + "learning_rate": 0.0001976396583361464, + "loss": 1.1662, + "step": 75 + }, + { + "epoch": 0.32, + "grad_norm": 0.2578125, + "learning_rate": 0.00019756677200175315, + "loss": 1.1222, + "step": 76 + }, + { + "epoch": 0.32421052631578945, + "grad_norm": 0.796875, + "learning_rate": 0.00019749279121818235, + "loss": 1.0329, + "step": 77 + }, + { + "epoch": 0.32842105263157895, + "grad_norm": 0.2490234375, + "learning_rate": 0.0001974177168153088, + "loss": 1.1199, + "step": 78 + }, + { + "epoch": 0.33263157894736844, + "grad_norm": 0.26171875, + "learning_rate": 0.00019734154963527495, + "loss": 1.4391, + "step": 79 + }, + { + "epoch": 0.3368421052631579, + "grad_norm": 0.30078125, + "learning_rate": 0.0001972642905324813, + "loss": 1.3924, + "step": 80 + }, + { + "epoch": 0.3410526315789474, + "grad_norm": 0.208984375, + "learning_rate": 0.00019718594037357713, + "loss": 1.1518, + "step": 81 + }, + { + "epoch": 0.3452631578947368, + "grad_norm": 0.953125, + "learning_rate": 0.0001971065000374504, + "loss": 1.2606, + "step": 82 + }, + { + "epoch": 0.3494736842105263, + "grad_norm": 0.306640625, + "learning_rate": 0.0001970259704152181, + "loss": 1.2293, + "step": 83 + }, + { + "epoch": 0.35368421052631577, + "grad_norm": 0.6328125, + "learning_rate": 0.0001969443524102163, + "loss": 0.9564, + "step": 84 + }, + { + "epoch": 0.35789473684210527, + "grad_norm": 0.279296875, + "learning_rate": 0.00019686164693798986, + "loss": 1.2023, + "step": 85 + }, + { + "epoch": 0.36210526315789476, + "grad_norm": 0.263671875, + "learning_rate": 0.0001967778549262822, + "loss": 0.9322, + "step": 86 + }, + { + "epoch": 0.3663157894736842, + "grad_norm": 0.51171875, + "learning_rate": 0.00019669297731502507, + "loss": 1.2589, + "step": 87 + }, + { + "epoch": 0.3705263157894737, + "grad_norm": 0.392578125, + "learning_rate": 0.00019660701505632772, + "loss": 1.3188, + "step": 88 + }, + { + "epoch": 0.37473684210526315, + "grad_norm": 0.51953125, + "learning_rate": 0.00019651996911446647, + "loss": 1.0415, + "step": 89 + }, + { + "epoch": 0.37894736842105264, + "grad_norm": 0.427734375, + "learning_rate": 0.0001964318404658737, + "loss": 1.467, + "step": 90 + }, + { + "epoch": 0.3831578947368421, + "grad_norm": 0.4453125, + "learning_rate": 0.00019634263009912705, + "loss": 1.5527, + "step": 91 + }, + { + "epoch": 0.3873684210526316, + "grad_norm": 0.30859375, + "learning_rate": 0.00019625233901493822, + "loss": 1.2043, + "step": 92 + }, + { + "epoch": 0.391578947368421, + "grad_norm": 0.279296875, + "learning_rate": 0.00019616096822614186, + "loss": 0.9677, + "step": 93 + }, + { + "epoch": 0.3957894736842105, + "grad_norm": 0.46484375, + "learning_rate": 0.000196068518757684, + "loss": 1.1942, + "step": 94 + }, + { + "epoch": 0.4, + "grad_norm": 0.435546875, + "learning_rate": 0.00019597499164661093, + "loss": 1.3481, + "step": 95 + }, + { + "epoch": 0.40421052631578946, + "grad_norm": 0.291015625, + "learning_rate": 0.00019588038794205703, + "loss": 1.2776, + "step": 96 + }, + { + "epoch": 0.40842105263157896, + "grad_norm": 0.3828125, + "learning_rate": 0.00019578470870523359, + "loss": 1.4808, + "step": 97 + }, + { + "epoch": 0.4126315789473684, + "grad_norm": 0.28125, + "learning_rate": 0.00019568795500941635, + "loss": 1.3385, + "step": 98 + }, + { + "epoch": 0.4168421052631579, + "grad_norm": 0.484375, + "learning_rate": 0.00019559012793993392, + "loss": 1.5787, + "step": 99 + }, + { + "epoch": 0.42105263157894735, + "grad_norm": 0.298828125, + "learning_rate": 0.00019549122859415538, + "loss": 1.2151, + "step": 100 + }, + { + "epoch": 0.42526315789473684, + "grad_norm": 0.65234375, + "learning_rate": 0.0001953912580814779, + "loss": 1.0332, + "step": 101 + }, + { + "epoch": 0.42947368421052634, + "grad_norm": 0.37890625, + "learning_rate": 0.00019529021752331453, + "loss": 1.2677, + "step": 102 + }, + { + "epoch": 0.4336842105263158, + "grad_norm": 0.265625, + "learning_rate": 0.0001951881080530814, + "loss": 1.0672, + "step": 103 + }, + { + "epoch": 0.4378947368421053, + "grad_norm": 0.498046875, + "learning_rate": 0.00019508493081618513, + "loss": 1.1736, + "step": 104 + }, + { + "epoch": 0.4421052631578947, + "grad_norm": 0.298828125, + "learning_rate": 0.00019498068697000994, + "loss": 1.1174, + "step": 105 + }, + { + "epoch": 0.4463157894736842, + "grad_norm": 0.169921875, + "learning_rate": 0.00019487537768390464, + "loss": 1.0712, + "step": 106 + }, + { + "epoch": 0.45052631578947366, + "grad_norm": 0.55078125, + "learning_rate": 0.0001947690041391696, + "loss": 1.4496, + "step": 107 + }, + { + "epoch": 0.45473684210526316, + "grad_norm": 0.47265625, + "learning_rate": 0.00019466156752904343, + "loss": 1.2413, + "step": 108 + }, + { + "epoch": 0.4589473684210526, + "grad_norm": 0.6015625, + "learning_rate": 0.0001945530690586896, + "loss": 1.0208, + "step": 109 + }, + { + "epoch": 0.4631578947368421, + "grad_norm": 0.35546875, + "learning_rate": 0.0001944435099451829, + "loss": 1.3997, + "step": 110 + }, + { + "epoch": 0.4673684210526316, + "grad_norm": 0.330078125, + "learning_rate": 0.00019433289141749586, + "loss": 1.2914, + "step": 111 + }, + { + "epoch": 0.47157894736842104, + "grad_norm": 0.63671875, + "learning_rate": 0.00019422121471648497, + "loss": 0.8134, + "step": 112 + }, + { + "epoch": 0.47578947368421054, + "grad_norm": 0.30859375, + "learning_rate": 0.00019410848109487658, + "loss": 1.1282, + "step": 113 + }, + { + "epoch": 0.48, + "grad_norm": 0.5625, + "learning_rate": 0.0001939946918172531, + "loss": 1.3254, + "step": 114 + }, + { + "epoch": 0.4842105263157895, + "grad_norm": 0.30859375, + "learning_rate": 0.00019387984816003867, + "loss": 1.6428, + "step": 115 + }, + { + "epoch": 0.4884210526315789, + "grad_norm": 0.33203125, + "learning_rate": 0.00019376395141148476, + "loss": 1.0688, + "step": 116 + }, + { + "epoch": 0.4926315789473684, + "grad_norm": 0.6953125, + "learning_rate": 0.00019364700287165604, + "loss": 1.0969, + "step": 117 + }, + { + "epoch": 0.4968421052631579, + "grad_norm": 1.296875, + "learning_rate": 0.00019352900385241536, + "loss": 1.5249, + "step": 118 + }, + { + "epoch": 0.5010526315789474, + "grad_norm": 0.2001953125, + "learning_rate": 0.00019340995567740938, + "loss": 1.2095, + "step": 119 + }, + { + "epoch": 0.5052631578947369, + "grad_norm": 0.53515625, + "learning_rate": 0.0001932898596820536, + "loss": 1.1249, + "step": 120 + }, + { + "epoch": 0.5052631578947369, + "eval_loss": 1.2166697978973389, + "eval_runtime": 25.8713, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.933, + "step": 120 + }, + { + "epoch": 0.5094736842105263, + "grad_norm": 0.423828125, + "learning_rate": 0.00019316871721351733, + "loss": 1.5017, + "step": 121 + }, + { + "epoch": 0.5136842105263157, + "grad_norm": 0.7421875, + "learning_rate": 0.0001930465296307087, + "loss": 1.0994, + "step": 122 + }, + { + "epoch": 0.5178947368421053, + "grad_norm": 0.21875, + "learning_rate": 0.0001929232983042592, + "loss": 1.0339, + "step": 123 + }, + { + "epoch": 0.5221052631578947, + "grad_norm": 0.6328125, + "learning_rate": 0.00019279902461650866, + "loss": 1.1317, + "step": 124 + }, + { + "epoch": 0.5263157894736842, + "grad_norm": 0.3984375, + "learning_rate": 0.00019267370996148934, + "loss": 0.9421, + "step": 125 + }, + { + "epoch": 0.5305263157894737, + "grad_norm": 0.396484375, + "learning_rate": 0.00019254735574491058, + "loss": 1.3029, + "step": 126 + }, + { + "epoch": 0.5347368421052632, + "grad_norm": 0.8359375, + "learning_rate": 0.00019241996338414296, + "loss": 1.3548, + "step": 127 + }, + { + "epoch": 0.5389473684210526, + "grad_norm": 0.84765625, + "learning_rate": 0.00019229153430820232, + "loss": 1.0865, + "step": 128 + }, + { + "epoch": 0.5431578947368421, + "grad_norm": 0.453125, + "learning_rate": 0.00019216206995773373, + "loss": 0.8151, + "step": 129 + }, + { + "epoch": 0.5473684210526316, + "grad_norm": 0.8046875, + "learning_rate": 0.0001920315717849956, + "loss": 1.3576, + "step": 130 + }, + { + "epoch": 0.5515789473684211, + "grad_norm": 0.82421875, + "learning_rate": 0.00019190004125384293, + "loss": 1.2439, + "step": 131 + }, + { + "epoch": 0.5557894736842105, + "grad_norm": 0.56640625, + "learning_rate": 0.0001917674798397113, + "loss": 0.9743, + "step": 132 + }, + { + "epoch": 0.56, + "grad_norm": 1.1015625, + "learning_rate": 0.0001916338890296001, + "loss": 1.3315, + "step": 133 + }, + { + "epoch": 0.5642105263157895, + "grad_norm": 0.333984375, + "learning_rate": 0.00019149927032205587, + "loss": 1.3268, + "step": 134 + }, + { + "epoch": 0.5684210526315789, + "grad_norm": 0.388671875, + "learning_rate": 0.00019136362522715562, + "loss": 1.4751, + "step": 135 + }, + { + "epoch": 0.5726315789473684, + "grad_norm": 0.2734375, + "learning_rate": 0.00019122695526648968, + "loss": 1.1219, + "step": 136 + }, + { + "epoch": 0.5768421052631579, + "grad_norm": 1.0703125, + "learning_rate": 0.00019108926197314486, + "loss": 1.4059, + "step": 137 + }, + { + "epoch": 0.5810526315789474, + "grad_norm": 0.33203125, + "learning_rate": 0.00019095054689168705, + "loss": 1.3282, + "step": 138 + }, + { + "epoch": 0.5852631578947368, + "grad_norm": 0.6171875, + "learning_rate": 0.00019081081157814404, + "loss": 1.6334, + "step": 139 + }, + { + "epoch": 0.5894736842105263, + "grad_norm": 0.359375, + "learning_rate": 0.00019067005759998797, + "loss": 1.159, + "step": 140 + }, + { + "epoch": 0.5936842105263158, + "grad_norm": 0.470703125, + "learning_rate": 0.00019052828653611786, + "loss": 1.3261, + "step": 141 + }, + { + "epoch": 0.5978947368421053, + "grad_norm": 0.26171875, + "learning_rate": 0.0001903854999768417, + "loss": 1.2041, + "step": 142 + }, + { + "epoch": 0.6021052631578947, + "grad_norm": 0.27734375, + "learning_rate": 0.00019024169952385885, + "loss": 1.1998, + "step": 143 + }, + { + "epoch": 0.6063157894736843, + "grad_norm": 0.56640625, + "learning_rate": 0.0001900968867902419, + "loss": 1.6427, + "step": 144 + }, + { + "epoch": 0.6105263157894737, + "grad_norm": 0.419921875, + "learning_rate": 0.00018995106340041873, + "loss": 1.1968, + "step": 145 + }, + { + "epoch": 0.6147368421052631, + "grad_norm": 0.671875, + "learning_rate": 0.00018980423099015402, + "loss": 0.8641, + "step": 146 + }, + { + "epoch": 0.6189473684210526, + "grad_norm": 0.333984375, + "learning_rate": 0.00018965639120653121, + "loss": 1.4518, + "step": 147 + }, + { + "epoch": 0.6231578947368421, + "grad_norm": 0.423828125, + "learning_rate": 0.00018950754570793384, + "loss": 1.3077, + "step": 148 + }, + { + "epoch": 0.6273684210526316, + "grad_norm": 0.85546875, + "learning_rate": 0.00018935769616402703, + "loss": 1.2077, + "step": 149 + }, + { + "epoch": 0.631578947368421, + "grad_norm": 0.224609375, + "learning_rate": 0.00018920684425573865, + "loss": 1.0868, + "step": 150 + }, + { + "epoch": 0.6357894736842106, + "grad_norm": 0.25390625, + "learning_rate": 0.00018905499167524062, + "loss": 1.2018, + "step": 151 + }, + { + "epoch": 0.64, + "grad_norm": 0.498046875, + "learning_rate": 0.00018890214012592975, + "loss": 1.1675, + "step": 152 + }, + { + "epoch": 0.6442105263157895, + "grad_norm": 0.2451171875, + "learning_rate": 0.00018874829132240874, + "loss": 1.2348, + "step": 153 + }, + { + "epoch": 0.6484210526315789, + "grad_norm": 0.33984375, + "learning_rate": 0.000188593446990467, + "loss": 1.3202, + "step": 154 + }, + { + "epoch": 0.6526315789473685, + "grad_norm": 1.0078125, + "learning_rate": 0.0001884376088670611, + "loss": 1.0171, + "step": 155 + }, + { + "epoch": 0.6568421052631579, + "grad_norm": 0.318359375, + "learning_rate": 0.00018828077870029552, + "loss": 1.219, + "step": 156 + }, + { + "epoch": 0.6610526315789473, + "grad_norm": 0.369140625, + "learning_rate": 0.00018812295824940285, + "loss": 0.8956, + "step": 157 + }, + { + "epoch": 0.6652631578947369, + "grad_norm": 0.328125, + "learning_rate": 0.00018796414928472417, + "loss": 0.9322, + "step": 158 + }, + { + "epoch": 0.6694736842105263, + "grad_norm": 0.2890625, + "learning_rate": 0.0001878043535876891, + "loss": 1.2062, + "step": 159 + }, + { + "epoch": 0.6736842105263158, + "grad_norm": 0.41015625, + "learning_rate": 0.0001876435729507959, + "loss": 1.3113, + "step": 160 + }, + { + "epoch": 0.6778947368421052, + "grad_norm": 0.53515625, + "learning_rate": 0.00018748180917759136, + "loss": 1.1539, + "step": 161 + }, + { + "epoch": 0.6821052631578948, + "grad_norm": 0.279296875, + "learning_rate": 0.0001873190640826505, + "loss": 1.3287, + "step": 162 + }, + { + "epoch": 0.6863157894736842, + "grad_norm": 0.298828125, + "learning_rate": 0.00018715533949155625, + "loss": 1.0616, + "step": 163 + }, + { + "epoch": 0.6905263157894737, + "grad_norm": 0.388671875, + "learning_rate": 0.00018699063724087904, + "loss": 1.0171, + "step": 164 + }, + { + "epoch": 0.6947368421052632, + "grad_norm": 0.32421875, + "learning_rate": 0.00018682495917815608, + "loss": 1.1113, + "step": 165 + }, + { + "epoch": 0.6989473684210527, + "grad_norm": 0.46875, + "learning_rate": 0.00018665830716187065, + "loss": 0.8351, + "step": 166 + }, + { + "epoch": 0.7031578947368421, + "grad_norm": 0.6015625, + "learning_rate": 0.0001864906830614314, + "loss": 1.2181, + "step": 167 + }, + { + "epoch": 0.7073684210526315, + "grad_norm": 0.71875, + "learning_rate": 0.0001863220887571512, + "loss": 1.1843, + "step": 168 + }, + { + "epoch": 0.7115789473684211, + "grad_norm": 0.51953125, + "learning_rate": 0.00018615252614022617, + "loss": 1.3809, + "step": 169 + }, + { + "epoch": 0.7157894736842105, + "grad_norm": 0.34375, + "learning_rate": 0.0001859819971127143, + "loss": 1.4288, + "step": 170 + }, + { + "epoch": 0.72, + "grad_norm": 0.41796875, + "learning_rate": 0.00018581050358751445, + "loss": 1.1449, + "step": 171 + }, + { + "epoch": 0.7242105263157895, + "grad_norm": 0.578125, + "learning_rate": 0.00018563804748834438, + "loss": 1.0703, + "step": 172 + }, + { + "epoch": 0.728421052631579, + "grad_norm": 0.43359375, + "learning_rate": 0.00018546463074971977, + "loss": 1.2093, + "step": 173 + }, + { + "epoch": 0.7326315789473684, + "grad_norm": 0.26171875, + "learning_rate": 0.000185290255316932, + "loss": 0.7346, + "step": 174 + }, + { + "epoch": 0.7368421052631579, + "grad_norm": 0.369140625, + "learning_rate": 0.00018511492314602656, + "loss": 1.2782, + "step": 175 + }, + { + "epoch": 0.7410526315789474, + "grad_norm": 0.56640625, + "learning_rate": 0.00018493863620378122, + "loss": 1.0844, + "step": 176 + }, + { + "epoch": 0.7452631578947368, + "grad_norm": 0.2470703125, + "learning_rate": 0.00018476139646768372, + "loss": 1.1771, + "step": 177 + }, + { + "epoch": 0.7494736842105263, + "grad_norm": 0.228515625, + "learning_rate": 0.00018458320592590975, + "loss": 1.2, + "step": 178 + }, + { + "epoch": 0.7536842105263157, + "grad_norm": 0.310546875, + "learning_rate": 0.00018440406657730058, + "loss": 1.2248, + "step": 179 + }, + { + "epoch": 0.7578947368421053, + "grad_norm": 0.263671875, + "learning_rate": 0.00018422398043134067, + "loss": 1.3675, + "step": 180 + }, + { + "epoch": 0.7578947368421053, + "eval_loss": 1.2129631042480469, + "eval_runtime": 25.9253, + "eval_samples_per_second": 3.857, + "eval_steps_per_second": 1.929, + "step": 180 + }, + { + "epoch": 0.7621052631578947, + "grad_norm": 0.2412109375, + "learning_rate": 0.0001840429495081352, + "loss": 1.2456, + "step": 181 + }, + { + "epoch": 0.7663157894736842, + "grad_norm": 0.25390625, + "learning_rate": 0.00018386097583838714, + "loss": 1.2617, + "step": 182 + }, + { + "epoch": 0.7705263157894737, + "grad_norm": 0.283203125, + "learning_rate": 0.00018367806146337483, + "loss": 1.2708, + "step": 183 + }, + { + "epoch": 0.7747368421052632, + "grad_norm": 0.318359375, + "learning_rate": 0.00018349420843492888, + "loss": 1.3771, + "step": 184 + }, + { + "epoch": 0.7789473684210526, + "grad_norm": 0.73046875, + "learning_rate": 0.00018330941881540915, + "loss": 1.3123, + "step": 185 + }, + { + "epoch": 0.783157894736842, + "grad_norm": 0.46875, + "learning_rate": 0.00018312369467768166, + "loss": 1.1952, + "step": 186 + }, + { + "epoch": 0.7873684210526316, + "grad_norm": 0.7578125, + "learning_rate": 0.00018293703810509537, + "loss": 1.426, + "step": 187 + }, + { + "epoch": 0.791578947368421, + "grad_norm": 0.33203125, + "learning_rate": 0.0001827494511914587, + "loss": 1.3205, + "step": 188 + }, + { + "epoch": 0.7957894736842105, + "grad_norm": 0.365234375, + "learning_rate": 0.00018256093604101617, + "loss": 1.2221, + "step": 189 + }, + { + "epoch": 0.8, + "grad_norm": 0.36328125, + "learning_rate": 0.0001823714947684247, + "loss": 1.4212, + "step": 190 + }, + { + "epoch": 0.8042105263157895, + "grad_norm": 0.349609375, + "learning_rate": 0.00018218112949872995, + "loss": 1.2183, + "step": 191 + }, + { + "epoch": 0.8084210526315789, + "grad_norm": 0.392578125, + "learning_rate": 0.00018198984236734246, + "loss": 0.9922, + "step": 192 + }, + { + "epoch": 0.8126315789473684, + "grad_norm": 0.373046875, + "learning_rate": 0.0001817976355200137, + "loss": 1.334, + "step": 193 + }, + { + "epoch": 0.8168421052631579, + "grad_norm": 0.30859375, + "learning_rate": 0.000181604511112812, + "loss": 1.1052, + "step": 194 + }, + { + "epoch": 0.8210526315789474, + "grad_norm": 0.349609375, + "learning_rate": 0.0001814104713120983, + "loss": 0.8112, + "step": 195 + }, + { + "epoch": 0.8252631578947368, + "grad_norm": 0.35546875, + "learning_rate": 0.000181215518294502, + "loss": 1.2507, + "step": 196 + }, + { + "epoch": 0.8294736842105264, + "grad_norm": 0.671875, + "learning_rate": 0.0001810196542468964, + "loss": 1.1228, + "step": 197 + }, + { + "epoch": 0.8336842105263158, + "grad_norm": 0.41015625, + "learning_rate": 0.00018082288136637422, + "loss": 1.3562, + "step": 198 + }, + { + "epoch": 0.8378947368421052, + "grad_norm": 0.302734375, + "learning_rate": 0.000180625201860223, + "loss": 0.7766, + "step": 199 + }, + { + "epoch": 0.8421052631578947, + "grad_norm": 0.322265625, + "learning_rate": 0.00018042661794590023, + "loss": 1.2885, + "step": 200 + }, + { + "epoch": 0.8463157894736842, + "grad_norm": 0.609375, + "learning_rate": 0.0001802271318510086, + "loss": 1.271, + "step": 201 + }, + { + "epoch": 0.8505263157894737, + "grad_norm": 0.341796875, + "learning_rate": 0.00018002674581327094, + "loss": 1.3777, + "step": 202 + }, + { + "epoch": 0.8547368421052631, + "grad_norm": 0.3828125, + "learning_rate": 0.00017982546208050515, + "loss": 1.32, + "step": 203 + }, + { + "epoch": 0.8589473684210527, + "grad_norm": 0.392578125, + "learning_rate": 0.00017962328291059888, + "loss": 1.1265, + "step": 204 + }, + { + "epoch": 0.8631578947368421, + "grad_norm": 0.73046875, + "learning_rate": 0.00017942021057148436, + "loss": 1.523, + "step": 205 + }, + { + "epoch": 0.8673684210526316, + "grad_norm": 1.046875, + "learning_rate": 0.00017921624734111292, + "loss": 1.0251, + "step": 206 + }, + { + "epoch": 0.871578947368421, + "grad_norm": 0.490234375, + "learning_rate": 0.0001790113955074294, + "loss": 1.2665, + "step": 207 + }, + { + "epoch": 0.8757894736842106, + "grad_norm": 0.6875, + "learning_rate": 0.0001788056573683464, + "loss": 1.2244, + "step": 208 + }, + { + "epoch": 0.88, + "grad_norm": 0.2177734375, + "learning_rate": 0.0001785990352317187, + "loss": 1.1621, + "step": 209 + }, + { + "epoch": 0.8842105263157894, + "grad_norm": 0.4921875, + "learning_rate": 0.00017839153141531718, + "loss": 1.3182, + "step": 210 + }, + { + "epoch": 0.888421052631579, + "grad_norm": 0.388671875, + "learning_rate": 0.000178183148246803, + "loss": 1.4508, + "step": 211 + }, + { + "epoch": 0.8926315789473684, + "grad_norm": 0.388671875, + "learning_rate": 0.00017797388806370132, + "loss": 1.1155, + "step": 212 + }, + { + "epoch": 0.8968421052631579, + "grad_norm": 0.416015625, + "learning_rate": 0.00017776375321337521, + "loss": 0.9191, + "step": 213 + }, + { + "epoch": 0.9010526315789473, + "grad_norm": 0.2431640625, + "learning_rate": 0.00017755274605299923, + "loss": 1.1577, + "step": 214 + }, + { + "epoch": 0.9052631578947369, + "grad_norm": 0.412109375, + "learning_rate": 0.00017734086894953305, + "loss": 1.372, + "step": 215 + }, + { + "epoch": 0.9094736842105263, + "grad_norm": 0.353515625, + "learning_rate": 0.00017712812427969485, + "loss": 1.1932, + "step": 216 + }, + { + "epoch": 0.9136842105263158, + "grad_norm": 0.291015625, + "learning_rate": 0.0001769145144299347, + "loss": 1.3167, + "step": 217 + }, + { + "epoch": 0.9178947368421052, + "grad_norm": 0.255859375, + "learning_rate": 0.00017670004179640774, + "loss": 1.2357, + "step": 218 + }, + { + "epoch": 0.9221052631578948, + "grad_norm": 0.5703125, + "learning_rate": 0.00017648470878494737, + "loss": 0.9556, + "step": 219 + }, + { + "epoch": 0.9263157894736842, + "grad_norm": 0.36328125, + "learning_rate": 0.0001762685178110382, + "loss": 1.2047, + "step": 220 + }, + { + "epoch": 0.9305263157894736, + "grad_norm": 0.271484375, + "learning_rate": 0.00017605147129978903, + "loss": 1.1935, + "step": 221 + }, + { + "epoch": 0.9347368421052632, + "grad_norm": 0.400390625, + "learning_rate": 0.0001758335716859055, + "loss": 1.3426, + "step": 222 + }, + { + "epoch": 0.9389473684210526, + "grad_norm": 0.73046875, + "learning_rate": 0.000175614821413663, + "loss": 1.5181, + "step": 223 + }, + { + "epoch": 0.9431578947368421, + "grad_norm": 0.3203125, + "learning_rate": 0.00017539522293687898, + "loss": 1.0962, + "step": 224 + }, + { + "epoch": 0.9473684210526315, + "grad_norm": 0.232421875, + "learning_rate": 0.00017517477871888568, + "loss": 1.3922, + "step": 225 + }, + { + "epoch": 0.9515789473684211, + "grad_norm": 0.5546875, + "learning_rate": 0.00017495349123250242, + "loss": 1.2535, + "step": 226 + }, + { + "epoch": 0.9557894736842105, + "grad_norm": 0.296875, + "learning_rate": 0.00017473136296000772, + "loss": 1.1987, + "step": 227 + }, + { + "epoch": 0.96, + "grad_norm": 0.77734375, + "learning_rate": 0.00017450839639311162, + "loss": 1.4301, + "step": 228 + }, + { + "epoch": 0.9642105263157895, + "grad_norm": 0.35546875, + "learning_rate": 0.00017428459403292767, + "loss": 1.3164, + "step": 229 + }, + { + "epoch": 0.968421052631579, + "grad_norm": 0.419921875, + "learning_rate": 0.00017405995838994494, + "loss": 1.1554, + "step": 230 + }, + { + "epoch": 0.9726315789473684, + "grad_norm": 0.3125, + "learning_rate": 0.00017383449198399964, + "loss": 1.3036, + "step": 231 + }, + { + "epoch": 0.9768421052631578, + "grad_norm": 0.267578125, + "learning_rate": 0.00017360819734424715, + "loss": 1.2797, + "step": 232 + }, + { + "epoch": 0.9810526315789474, + "grad_norm": 0.412109375, + "learning_rate": 0.00017338107700913354, + "loss": 1.1202, + "step": 233 + }, + { + "epoch": 0.9852631578947368, + "grad_norm": 0.365234375, + "learning_rate": 0.0001731531335263669, + "loss": 0.8187, + "step": 234 + }, + { + "epoch": 0.9894736842105263, + "grad_norm": 0.5546875, + "learning_rate": 0.0001729243694528891, + "loss": 1.1702, + "step": 235 + }, + { + "epoch": 0.9936842105263158, + "grad_norm": 0.5546875, + "learning_rate": 0.00017269478735484683, + "loss": 1.1406, + "step": 236 + }, + { + "epoch": 0.9978947368421053, + "grad_norm": 0.3359375, + "learning_rate": 0.00017246438980756293, + "loss": 0.9633, + "step": 237 + }, + { + "epoch": 1.0021052631578948, + "grad_norm": 0.359375, + "learning_rate": 0.00017223317939550753, + "loss": 1.2181, + "step": 238 + }, + { + "epoch": 1.0063157894736843, + "grad_norm": 0.2138671875, + "learning_rate": 0.00017200115871226904, + "loss": 1.2035, + "step": 239 + }, + { + "epoch": 1.0105263157894737, + "grad_norm": 0.3203125, + "learning_rate": 0.00017176833036052495, + "loss": 1.3449, + "step": 240 + }, + { + "epoch": 1.0105263157894737, + "eval_loss": 1.1994891166687012, + "eval_runtime": 25.9787, + "eval_samples_per_second": 3.849, + "eval_steps_per_second": 1.925, + "step": 240 + }, + { + "epoch": 1.0147368421052632, + "grad_norm": 0.375, + "learning_rate": 0.00017153469695201277, + "loss": 1.0695, + "step": 241 + }, + { + "epoch": 1.0189473684210526, + "grad_norm": 0.578125, + "learning_rate": 0.0001713002611075007, + "loss": 0.8898, + "step": 242 + }, + { + "epoch": 1.023157894736842, + "grad_norm": 0.203125, + "learning_rate": 0.0001710650254567582, + "loss": 1.087, + "step": 243 + }, + { + "epoch": 1.0273684210526315, + "grad_norm": 0.3828125, + "learning_rate": 0.0001708289926385265, + "loss": 1.1739, + "step": 244 + }, + { + "epoch": 1.0315789473684212, + "grad_norm": 0.3515625, + "learning_rate": 0.000170592165300489, + "loss": 1.0413, + "step": 245 + }, + { + "epoch": 1.0357894736842106, + "grad_norm": 0.376953125, + "learning_rate": 0.0001703545460992416, + "loss": 1.0745, + "step": 246 + }, + { + "epoch": 1.04, + "grad_norm": 0.25, + "learning_rate": 0.00017011613770026292, + "loss": 1.2494, + "step": 247 + }, + { + "epoch": 1.0442105263157895, + "grad_norm": 0.419921875, + "learning_rate": 0.00016987694277788417, + "loss": 0.7568, + "step": 248 + }, + { + "epoch": 1.048421052631579, + "grad_norm": 0.6015625, + "learning_rate": 0.0001696369640152596, + "loss": 1.2675, + "step": 249 + }, + { + "epoch": 1.0526315789473684, + "grad_norm": 0.67578125, + "learning_rate": 0.0001693962041043359, + "loss": 1.0456, + "step": 250 + }, + { + "epoch": 1.0568421052631578, + "grad_norm": 0.5234375, + "learning_rate": 0.00016915466574582246, + "loss": 1.1145, + "step": 251 + }, + { + "epoch": 1.0610526315789475, + "grad_norm": 0.279296875, + "learning_rate": 0.00016891235164916065, + "loss": 1.1659, + "step": 252 + }, + { + "epoch": 1.065263157894737, + "grad_norm": 0.369140625, + "learning_rate": 0.0001686692645324938, + "loss": 0.9993, + "step": 253 + }, + { + "epoch": 1.0694736842105264, + "grad_norm": 0.78125, + "learning_rate": 0.00016842540712263637, + "loss": 1.2847, + "step": 254 + }, + { + "epoch": 1.0736842105263158, + "grad_norm": 0.48046875, + "learning_rate": 0.0001681807821550438, + "loss": 1.1483, + "step": 255 + }, + { + "epoch": 1.0778947368421052, + "grad_norm": 0.73046875, + "learning_rate": 0.00016793539237378128, + "loss": 0.7895, + "step": 256 + }, + { + "epoch": 1.0821052631578947, + "grad_norm": 0.30859375, + "learning_rate": 0.00016768924053149338, + "loss": 1.2321, + "step": 257 + }, + { + "epoch": 1.0863157894736841, + "grad_norm": 0.291015625, + "learning_rate": 0.00016744232938937308, + "loss": 1.0098, + "step": 258 + }, + { + "epoch": 1.0905263157894738, + "grad_norm": 0.470703125, + "learning_rate": 0.00016719466171713073, + "loss": 1.3362, + "step": 259 + }, + { + "epoch": 1.0947368421052632, + "grad_norm": 0.2890625, + "learning_rate": 0.0001669462402929629, + "loss": 1.2487, + "step": 260 + }, + { + "epoch": 1.0989473684210527, + "grad_norm": 0.39453125, + "learning_rate": 0.00016669706790352148, + "loss": 1.1732, + "step": 261 + }, + { + "epoch": 1.1031578947368421, + "grad_norm": 0.28125, + "learning_rate": 0.00016644714734388217, + "loss": 1.1493, + "step": 262 + }, + { + "epoch": 1.1073684210526316, + "grad_norm": 0.314453125, + "learning_rate": 0.0001661964814175133, + "loss": 1.0521, + "step": 263 + }, + { + "epoch": 1.111578947368421, + "grad_norm": 0.53125, + "learning_rate": 0.00016594507293624425, + "loss": 1.07, + "step": 264 + }, + { + "epoch": 1.1157894736842104, + "grad_norm": 0.2890625, + "learning_rate": 0.00016569292472023392, + "loss": 1.1393, + "step": 265 + }, + { + "epoch": 1.12, + "grad_norm": 0.390625, + "learning_rate": 0.00016544003959793925, + "loss": 1.2059, + "step": 266 + }, + { + "epoch": 1.1242105263157895, + "grad_norm": 0.494140625, + "learning_rate": 0.0001651864204060833, + "loss": 0.9832, + "step": 267 + }, + { + "epoch": 1.128421052631579, + "grad_norm": 0.376953125, + "learning_rate": 0.00016493206998962354, + "loss": 0.9154, + "step": 268 + }, + { + "epoch": 1.1326315789473684, + "grad_norm": 0.39453125, + "learning_rate": 0.00016467699120171987, + "loss": 1.0839, + "step": 269 + }, + { + "epoch": 1.1368421052631579, + "grad_norm": 0.94140625, + "learning_rate": 0.0001644211869037027, + "loss": 1.1298, + "step": 270 + }, + { + "epoch": 1.1410526315789473, + "grad_norm": 0.341796875, + "learning_rate": 0.00016416465996504077, + "loss": 1.131, + "step": 271 + }, + { + "epoch": 1.1452631578947368, + "grad_norm": 0.90234375, + "learning_rate": 0.00016390741326330907, + "loss": 1.0148, + "step": 272 + }, + { + "epoch": 1.1494736842105264, + "grad_norm": 0.35546875, + "learning_rate": 0.00016364944968415632, + "loss": 0.788, + "step": 273 + }, + { + "epoch": 1.1536842105263159, + "grad_norm": 0.294921875, + "learning_rate": 0.00016339077212127294, + "loss": 0.9585, + "step": 274 + }, + { + "epoch": 1.1578947368421053, + "grad_norm": 0.56640625, + "learning_rate": 0.00016313138347635837, + "loss": 0.9937, + "step": 275 + }, + { + "epoch": 1.1621052631578948, + "grad_norm": 0.251953125, + "learning_rate": 0.0001628712866590885, + "loss": 1.069, + "step": 276 + }, + { + "epoch": 1.1663157894736842, + "grad_norm": 0.384765625, + "learning_rate": 0.00016261048458708318, + "loss": 1.2734, + "step": 277 + }, + { + "epoch": 1.1705263157894736, + "grad_norm": 0.66015625, + "learning_rate": 0.00016234898018587337, + "loss": 1.1955, + "step": 278 + }, + { + "epoch": 1.174736842105263, + "grad_norm": 0.3984375, + "learning_rate": 0.00016208677638886835, + "loss": 0.9949, + "step": 279 + }, + { + "epoch": 1.1789473684210527, + "grad_norm": 0.33984375, + "learning_rate": 0.00016182387613732291, + "loss": 1.1192, + "step": 280 + }, + { + "epoch": 1.1831578947368422, + "grad_norm": 0.4765625, + "learning_rate": 0.00016156028238030422, + "loss": 1.165, + "step": 281 + }, + { + "epoch": 1.1873684210526316, + "grad_norm": 0.51171875, + "learning_rate": 0.00016129599807465875, + "loss": 0.9109, + "step": 282 + }, + { + "epoch": 1.191578947368421, + "grad_norm": 0.5546875, + "learning_rate": 0.00016103102618497922, + "loss": 0.9441, + "step": 283 + }, + { + "epoch": 1.1957894736842105, + "grad_norm": 0.7109375, + "learning_rate": 0.0001607653696835713, + "loss": 1.3679, + "step": 284 + }, + { + "epoch": 1.2, + "grad_norm": 0.3671875, + "learning_rate": 0.0001604990315504202, + "loss": 0.8203, + "step": 285 + }, + { + "epoch": 1.2042105263157894, + "grad_norm": 0.51171875, + "learning_rate": 0.00016023201477315731, + "loss": 1.052, + "step": 286 + }, + { + "epoch": 1.208421052631579, + "grad_norm": 0.72265625, + "learning_rate": 0.00015996432234702666, + "loss": 1.281, + "step": 287 + }, + { + "epoch": 1.2126315789473685, + "grad_norm": 0.44921875, + "learning_rate": 0.0001596959572748514, + "loss": 1.1851, + "step": 288 + }, + { + "epoch": 1.216842105263158, + "grad_norm": 1.3515625, + "learning_rate": 0.0001594269225669999, + "loss": 1.1378, + "step": 289 + }, + { + "epoch": 1.2210526315789474, + "grad_norm": 0.337890625, + "learning_rate": 0.00015915722124135227, + "loss": 0.7495, + "step": 290 + }, + { + "epoch": 1.2252631578947368, + "grad_norm": 0.45703125, + "learning_rate": 0.00015888685632326627, + "loss": 1.0089, + "step": 291 + }, + { + "epoch": 1.2294736842105263, + "grad_norm": 0.470703125, + "learning_rate": 0.00015861583084554349, + "loss": 0.7145, + "step": 292 + }, + { + "epoch": 1.2336842105263157, + "grad_norm": 0.4375, + "learning_rate": 0.00015834414784839532, + "loss": 1.1728, + "step": 293 + }, + { + "epoch": 1.2378947368421054, + "grad_norm": 1.5703125, + "learning_rate": 0.0001580718103794089, + "loss": 1.351, + "step": 294 + }, + { + "epoch": 1.2421052631578948, + "grad_norm": 0.6171875, + "learning_rate": 0.0001577988214935127, + "loss": 1.0116, + "step": 295 + }, + { + "epoch": 1.2463157894736843, + "grad_norm": 0.392578125, + "learning_rate": 0.00015752518425294257, + "loss": 1.1591, + "step": 296 + }, + { + "epoch": 1.2505263157894737, + "grad_norm": 0.44140625, + "learning_rate": 0.0001572509017272072, + "loss": 1.0848, + "step": 297 + }, + { + "epoch": 1.2547368421052632, + "grad_norm": 0.43359375, + "learning_rate": 0.00015697597699305366, + "loss": 1.0024, + "step": 298 + }, + { + "epoch": 1.2589473684210526, + "grad_norm": 0.470703125, + "learning_rate": 0.00015670041313443304, + "loss": 1.072, + "step": 299 + }, + { + "epoch": 1.263157894736842, + "grad_norm": 0.34375, + "learning_rate": 0.00015642421324246568, + "loss": 1.1825, + "step": 300 + }, + { + "epoch": 1.263157894736842, + "eval_loss": 1.207429051399231, + "eval_runtime": 25.9189, + "eval_samples_per_second": 3.858, + "eval_steps_per_second": 1.929, + "step": 300 + }, + { + "epoch": 1.2673684210526317, + "grad_norm": 1.0234375, + "learning_rate": 0.00015614738041540668, + "loss": 0.8984, + "step": 301 + }, + { + "epoch": 1.271578947368421, + "grad_norm": 0.29296875, + "learning_rate": 0.00015586991775861102, + "loss": 0.9569, + "step": 302 + }, + { + "epoch": 1.2757894736842106, + "grad_norm": 0.59375, + "learning_rate": 0.00015559182838449873, + "loss": 1.177, + "step": 303 + }, + { + "epoch": 1.28, + "grad_norm": 0.416015625, + "learning_rate": 0.00015531311541251995, + "loss": 0.9957, + "step": 304 + }, + { + "epoch": 1.2842105263157895, + "grad_norm": 0.47265625, + "learning_rate": 0.00015503378196912013, + "loss": 0.8083, + "step": 305 + }, + { + "epoch": 1.288421052631579, + "grad_norm": 0.7734375, + "learning_rate": 0.00015475383118770472, + "loss": 1.0463, + "step": 306 + }, + { + "epoch": 1.2926315789473684, + "grad_norm": 0.30859375, + "learning_rate": 0.00015447326620860415, + "loss": 1.0709, + "step": 307 + }, + { + "epoch": 1.296842105263158, + "grad_norm": 0.35546875, + "learning_rate": 0.00015419209017903852, + "loss": 1.2547, + "step": 308 + }, + { + "epoch": 1.3010526315789472, + "grad_norm": 0.3671875, + "learning_rate": 0.00015391030625308249, + "loss": 0.8027, + "step": 309 + }, + { + "epoch": 1.305263157894737, + "grad_norm": 0.28515625, + "learning_rate": 0.0001536279175916296, + "loss": 1.0735, + "step": 310 + }, + { + "epoch": 1.3094736842105263, + "grad_norm": 0.625, + "learning_rate": 0.00015334492736235705, + "loss": 1.0046, + "step": 311 + }, + { + "epoch": 1.3136842105263158, + "grad_norm": 0.83203125, + "learning_rate": 0.0001530613387396901, + "loss": 1.1327, + "step": 312 + }, + { + "epoch": 1.3178947368421052, + "grad_norm": 0.515625, + "learning_rate": 0.00015277715490476641, + "loss": 0.9917, + "step": 313 + }, + { + "epoch": 1.3221052631578947, + "grad_norm": 0.359375, + "learning_rate": 0.0001524923790454004, + "loss": 1.2532, + "step": 314 + }, + { + "epoch": 1.3263157894736843, + "grad_norm": 0.306640625, + "learning_rate": 0.0001522070143560476, + "loss": 0.9906, + "step": 315 + }, + { + "epoch": 1.3305263157894736, + "grad_norm": 0.33984375, + "learning_rate": 0.00015192106403776848, + "loss": 1.0958, + "step": 316 + }, + { + "epoch": 1.3347368421052632, + "grad_norm": 0.33203125, + "learning_rate": 0.000151634531298193, + "loss": 1.1674, + "step": 317 + }, + { + "epoch": 1.3389473684210527, + "grad_norm": 0.45703125, + "learning_rate": 0.0001513474193514842, + "loss": 0.9189, + "step": 318 + }, + { + "epoch": 1.343157894736842, + "grad_norm": 0.55078125, + "learning_rate": 0.0001510597314183025, + "loss": 1.2956, + "step": 319 + }, + { + "epoch": 1.3473684210526315, + "grad_norm": 0.41015625, + "learning_rate": 0.00015077147072576933, + "loss": 1.3053, + "step": 320 + }, + { + "epoch": 1.351578947368421, + "grad_norm": 0.421875, + "learning_rate": 0.00015048264050743096, + "loss": 1.26, + "step": 321 + }, + { + "epoch": 1.3557894736842107, + "grad_norm": 0.318359375, + "learning_rate": 0.00015019324400322243, + "loss": 1.1314, + "step": 322 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 0.33203125, + "learning_rate": 0.00014990328445943092, + "loss": 1.1536, + "step": 323 + }, + { + "epoch": 1.3642105263157895, + "grad_norm": 0.365234375, + "learning_rate": 0.00014961276512865954, + "loss": 0.8678, + "step": 324 + }, + { + "epoch": 1.368421052631579, + "grad_norm": 0.56640625, + "learning_rate": 0.00014932168926979074, + "loss": 1.0508, + "step": 325 + }, + { + "epoch": 1.3726315789473684, + "grad_norm": 0.333984375, + "learning_rate": 0.00014903006014794983, + "loss": 0.9275, + "step": 326 + }, + { + "epoch": 1.3768421052631579, + "grad_norm": 0.61328125, + "learning_rate": 0.00014873788103446825, + "loss": 1.0225, + "step": 327 + }, + { + "epoch": 1.3810526315789473, + "grad_norm": 0.357421875, + "learning_rate": 0.00014844515520684703, + "loss": 0.9955, + "step": 328 + }, + { + "epoch": 1.385263157894737, + "grad_norm": 0.6328125, + "learning_rate": 0.00014815188594871983, + "loss": 1.2857, + "step": 329 + }, + { + "epoch": 1.3894736842105262, + "grad_norm": 0.5078125, + "learning_rate": 0.00014785807654981627, + "loss": 1.0598, + "step": 330 + }, + { + "epoch": 1.3936842105263159, + "grad_norm": 0.96484375, + "learning_rate": 0.00014756373030592497, + "loss": 1.1031, + "step": 331 + }, + { + "epoch": 1.3978947368421053, + "grad_norm": 0.380859375, + "learning_rate": 0.00014726885051885653, + "loss": 1.3287, + "step": 332 + }, + { + "epoch": 1.4021052631578947, + "grad_norm": 0.5859375, + "learning_rate": 0.00014697344049640658, + "loss": 1.1952, + "step": 333 + }, + { + "epoch": 1.4063157894736842, + "grad_norm": 0.453125, + "learning_rate": 0.0001466775035523186, + "loss": 1.083, + "step": 334 + }, + { + "epoch": 1.4105263157894736, + "grad_norm": 0.37890625, + "learning_rate": 0.00014638104300624685, + "loss": 0.8393, + "step": 335 + }, + { + "epoch": 1.4147368421052633, + "grad_norm": 0.9921875, + "learning_rate": 0.00014608406218371894, + "loss": 1.3947, + "step": 336 + }, + { + "epoch": 1.4189473684210525, + "grad_norm": 0.357421875, + "learning_rate": 0.0001457865644160987, + "loss": 1.028, + "step": 337 + }, + { + "epoch": 1.4231578947368422, + "grad_norm": 0.671875, + "learning_rate": 0.00014548855304054886, + "loss": 1.1156, + "step": 338 + }, + { + "epoch": 1.4273684210526316, + "grad_norm": 0.6953125, + "learning_rate": 0.00014519003139999337, + "loss": 1.1537, + "step": 339 + }, + { + "epoch": 1.431578947368421, + "grad_norm": 0.408203125, + "learning_rate": 0.00014489100284308017, + "loss": 1.207, + "step": 340 + }, + { + "epoch": 1.4357894736842105, + "grad_norm": 0.56640625, + "learning_rate": 0.00014459147072414335, + "loss": 1.125, + "step": 341 + }, + { + "epoch": 1.44, + "grad_norm": 0.423828125, + "learning_rate": 0.00014429143840316585, + "loss": 1.0774, + "step": 342 + }, + { + "epoch": 1.4442105263157896, + "grad_norm": 1.0859375, + "learning_rate": 0.0001439909092457415, + "loss": 1.3349, + "step": 343 + }, + { + "epoch": 1.4484210526315788, + "grad_norm": 0.55078125, + "learning_rate": 0.00014368988662303732, + "loss": 0.6138, + "step": 344 + }, + { + "epoch": 1.4526315789473685, + "grad_norm": 0.330078125, + "learning_rate": 0.00014338837391175582, + "loss": 1.1961, + "step": 345 + }, + { + "epoch": 1.456842105263158, + "grad_norm": 0.54296875, + "learning_rate": 0.00014308637449409706, + "loss": 1.067, + "step": 346 + }, + { + "epoch": 1.4610526315789474, + "grad_norm": 0.294921875, + "learning_rate": 0.0001427838917577206, + "loss": 1.0033, + "step": 347 + }, + { + "epoch": 1.4652631578947368, + "grad_norm": 0.494140625, + "learning_rate": 0.00014248092909570774, + "loss": 1.011, + "step": 348 + }, + { + "epoch": 1.4694736842105263, + "grad_norm": 0.75, + "learning_rate": 0.00014217748990652318, + "loss": 1.3504, + "step": 349 + }, + { + "epoch": 1.4736842105263157, + "grad_norm": 0.41796875, + "learning_rate": 0.00014187357759397714, + "loss": 1.0421, + "step": 350 + }, + { + "epoch": 1.4778947368421052, + "grad_norm": 0.73828125, + "learning_rate": 0.00014156919556718698, + "loss": 1.1829, + "step": 351 + }, + { + "epoch": 1.4821052631578948, + "grad_norm": 0.31640625, + "learning_rate": 0.00014126434724053913, + "loss": 1.1985, + "step": 352 + }, + { + "epoch": 1.4863157894736843, + "grad_norm": 0.41015625, + "learning_rate": 0.00014095903603365066, + "loss": 1.1609, + "step": 353 + }, + { + "epoch": 1.4905263157894737, + "grad_norm": 0.345703125, + "learning_rate": 0.00014065326537133094, + "loss": 1.2552, + "step": 354 + }, + { + "epoch": 1.4947368421052631, + "grad_norm": 1.2265625, + "learning_rate": 0.0001403470386835434, + "loss": 0.9655, + "step": 355 + }, + { + "epoch": 1.4989473684210526, + "grad_norm": 0.46875, + "learning_rate": 0.0001400403594053667, + "loss": 0.9709, + "step": 356 + }, + { + "epoch": 1.5031578947368422, + "grad_norm": 0.703125, + "learning_rate": 0.00013973323097695657, + "loss": 1.048, + "step": 357 + }, + { + "epoch": 1.5073684210526315, + "grad_norm": 0.494140625, + "learning_rate": 0.00013942565684350698, + "loss": 1.0853, + "step": 358 + }, + { + "epoch": 1.5115789473684211, + "grad_norm": 0.9140625, + "learning_rate": 0.00013911764045521159, + "loss": 1.1191, + "step": 359 + }, + { + "epoch": 1.5157894736842106, + "grad_norm": 0.478515625, + "learning_rate": 0.00013880918526722497, + "loss": 0.9782, + "step": 360 + }, + { + "epoch": 1.5157894736842106, + "eval_loss": 1.2060081958770752, + "eval_runtime": 25.9425, + "eval_samples_per_second": 3.855, + "eval_steps_per_second": 1.927, + "step": 360 + }, + { + "epoch": 1.52, + "grad_norm": 0.361328125, + "learning_rate": 0.00013850029473962402, + "loss": 0.9822, + "step": 361 + }, + { + "epoch": 1.5242105263157895, + "grad_norm": 0.44921875, + "learning_rate": 0.00013819097233736888, + "loss": 0.957, + "step": 362 + }, + { + "epoch": 1.528421052631579, + "grad_norm": 0.478515625, + "learning_rate": 0.0001378812215302644, + "loss": 0.9774, + "step": 363 + }, + { + "epoch": 1.5326315789473686, + "grad_norm": 0.35546875, + "learning_rate": 0.00013757104579292082, + "loss": 1.2854, + "step": 364 + }, + { + "epoch": 1.5368421052631578, + "grad_norm": 0.671875, + "learning_rate": 0.00013726044860471522, + "loss": 0.9209, + "step": 365 + }, + { + "epoch": 1.5410526315789475, + "grad_norm": 0.400390625, + "learning_rate": 0.00013694943344975212, + "loss": 0.9489, + "step": 366 + }, + { + "epoch": 1.545263157894737, + "grad_norm": 0.625, + "learning_rate": 0.00013663800381682464, + "loss": 1.2956, + "step": 367 + }, + { + "epoch": 1.5494736842105263, + "grad_norm": 0.3203125, + "learning_rate": 0.00013632616319937522, + "loss": 1.0642, + "step": 368 + }, + { + "epoch": 1.5536842105263158, + "grad_norm": 0.296875, + "learning_rate": 0.0001360139150954566, + "loss": 1.034, + "step": 369 + }, + { + "epoch": 1.5578947368421052, + "grad_norm": 0.5234375, + "learning_rate": 0.00013570126300769232, + "loss": 0.8314, + "step": 370 + }, + { + "epoch": 1.5621052631578949, + "grad_norm": 0.279296875, + "learning_rate": 0.00013538821044323772, + "loss": 1.1934, + "step": 371 + }, + { + "epoch": 1.566315789473684, + "grad_norm": 0.353515625, + "learning_rate": 0.0001350747609137404, + "loss": 1.2661, + "step": 372 + }, + { + "epoch": 1.5705263157894738, + "grad_norm": 0.27734375, + "learning_rate": 0.00013476091793530092, + "loss": 0.9746, + "step": 373 + }, + { + "epoch": 1.5747368421052632, + "grad_norm": 0.49609375, + "learning_rate": 0.0001344466850284333, + "loss": 1.0562, + "step": 374 + }, + { + "epoch": 1.5789473684210527, + "grad_norm": 0.27734375, + "learning_rate": 0.00013413206571802558, + "loss": 0.8612, + "step": 375 + }, + { + "epoch": 1.583157894736842, + "grad_norm": 0.51953125, + "learning_rate": 0.00013381706353330014, + "loss": 0.7912, + "step": 376 + }, + { + "epoch": 1.5873684210526315, + "grad_norm": 0.45703125, + "learning_rate": 0.00013350168200777448, + "loss": 1.0244, + "step": 377 + }, + { + "epoch": 1.5915789473684212, + "grad_norm": 0.5703125, + "learning_rate": 0.0001331859246792211, + "loss": 1.1267, + "step": 378 + }, + { + "epoch": 1.5957894736842104, + "grad_norm": 0.48828125, + "learning_rate": 0.00013286979508962815, + "loss": 1.0055, + "step": 379 + }, + { + "epoch": 1.6, + "grad_norm": 0.310546875, + "learning_rate": 0.0001325532967851596, + "loss": 1.2348, + "step": 380 + }, + { + "epoch": 1.6042105263157893, + "grad_norm": 0.369140625, + "learning_rate": 0.00013223643331611537, + "loss": 0.991, + "step": 381 + }, + { + "epoch": 1.608421052631579, + "grad_norm": 0.51171875, + "learning_rate": 0.00013191920823689177, + "loss": 1.0261, + "step": 382 + }, + { + "epoch": 1.6126315789473684, + "grad_norm": 0.90234375, + "learning_rate": 0.00013160162510594123, + "loss": 1.0752, + "step": 383 + }, + { + "epoch": 1.6168421052631579, + "grad_norm": 1.1796875, + "learning_rate": 0.00013128368748573273, + "loss": 1.1808, + "step": 384 + }, + { + "epoch": 1.6210526315789475, + "grad_norm": 0.40625, + "learning_rate": 0.0001309653989427117, + "loss": 0.8969, + "step": 385 + }, + { + "epoch": 1.6252631578947367, + "grad_norm": 0.53515625, + "learning_rate": 0.00013064676304726, + "loss": 1.1139, + "step": 386 + }, + { + "epoch": 1.6294736842105264, + "grad_norm": 0.478515625, + "learning_rate": 0.00013032778337365592, + "loss": 1.2781, + "step": 387 + }, + { + "epoch": 1.6336842105263156, + "grad_norm": 0.66796875, + "learning_rate": 0.0001300084635000341, + "loss": 1.1434, + "step": 388 + }, + { + "epoch": 1.6378947368421053, + "grad_norm": 0.33984375, + "learning_rate": 0.00012968880700834517, + "loss": 0.7428, + "step": 389 + }, + { + "epoch": 1.6421052631578947, + "grad_norm": 0.357421875, + "learning_rate": 0.000129368817484316, + "loss": 1.256, + "step": 390 + }, + { + "epoch": 1.6463157894736842, + "grad_norm": 0.384765625, + "learning_rate": 0.00012904849851740906, + "loss": 1.1516, + "step": 391 + }, + { + "epoch": 1.6505263157894738, + "grad_norm": 0.62890625, + "learning_rate": 0.0001287278537007824, + "loss": 1.1683, + "step": 392 + }, + { + "epoch": 1.654736842105263, + "grad_norm": 0.4140625, + "learning_rate": 0.0001284068866312492, + "loss": 0.9682, + "step": 393 + }, + { + "epoch": 1.6589473684210527, + "grad_norm": 0.36328125, + "learning_rate": 0.00012808560090923758, + "loss": 1.2109, + "step": 394 + }, + { + "epoch": 1.663157894736842, + "grad_norm": 0.3125, + "learning_rate": 0.00012776400013875006, + "loss": 1.1139, + "step": 395 + }, + { + "epoch": 1.6673684210526316, + "grad_norm": 0.453125, + "learning_rate": 0.00012744208792732324, + "loss": 1.0977, + "step": 396 + }, + { + "epoch": 1.671578947368421, + "grad_norm": 0.44140625, + "learning_rate": 0.00012711986788598725, + "loss": 1.1872, + "step": 397 + }, + { + "epoch": 1.6757894736842105, + "grad_norm": 0.59765625, + "learning_rate": 0.00012679734362922528, + "loss": 1.0773, + "step": 398 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 0.265625, + "learning_rate": 0.00012647451877493307, + "loss": 0.6203, + "step": 399 + }, + { + "epoch": 1.6842105263157894, + "grad_norm": 0.5078125, + "learning_rate": 0.00012615139694437835, + "loss": 1.1557, + "step": 400 + }, + { + "epoch": 1.688421052631579, + "grad_norm": 0.578125, + "learning_rate": 0.00012582798176215997, + "loss": 1.1022, + "step": 401 + }, + { + "epoch": 1.6926315789473683, + "grad_norm": 0.4609375, + "learning_rate": 0.00012550427685616765, + "loss": 1.193, + "step": 402 + }, + { + "epoch": 1.696842105263158, + "grad_norm": 0.455078125, + "learning_rate": 0.00012518028585754094, + "loss": 0.8758, + "step": 403 + }, + { + "epoch": 1.7010526315789474, + "grad_norm": 0.51171875, + "learning_rate": 0.00012485601240062869, + "loss": 0.9792, + "step": 404 + }, + { + "epoch": 1.7052631578947368, + "grad_norm": 0.3046875, + "learning_rate": 0.00012453146012294812, + "loss": 1.1347, + "step": 405 + }, + { + "epoch": 1.7094736842105265, + "grad_norm": 0.439453125, + "learning_rate": 0.00012420663266514417, + "loss": 1.1692, + "step": 406 + }, + { + "epoch": 1.7136842105263157, + "grad_norm": 0.33203125, + "learning_rate": 0.00012388153367094858, + "loss": 0.9904, + "step": 407 + }, + { + "epoch": 1.7178947368421054, + "grad_norm": 0.62109375, + "learning_rate": 0.0001235561667871391, + "loss": 0.8854, + "step": 408 + }, + { + "epoch": 1.7221052631578946, + "grad_norm": 0.52734375, + "learning_rate": 0.00012323053566349834, + "loss": 1.3143, + "step": 409 + }, + { + "epoch": 1.7263157894736842, + "grad_norm": 0.390625, + "learning_rate": 0.0001229046439527732, + "loss": 1.0957, + "step": 410 + }, + { + "epoch": 1.7305263157894737, + "grad_norm": 0.494140625, + "learning_rate": 0.00012257849531063352, + "loss": 1.0458, + "step": 411 + }, + { + "epoch": 1.7347368421052631, + "grad_norm": 0.34375, + "learning_rate": 0.00012225209339563145, + "loss": 1.3262, + "step": 412 + }, + { + "epoch": 1.7389473684210528, + "grad_norm": 0.32421875, + "learning_rate": 0.00012192544186916006, + "loss": 1.1506, + "step": 413 + }, + { + "epoch": 1.743157894736842, + "grad_norm": 0.671875, + "learning_rate": 0.00012159854439541245, + "loss": 1.035, + "step": 414 + }, + { + "epoch": 1.7473684210526317, + "grad_norm": 0.4453125, + "learning_rate": 0.00012127140464134066, + "loss": 1.3485, + "step": 415 + }, + { + "epoch": 1.751578947368421, + "grad_norm": 0.439453125, + "learning_rate": 0.00012094402627661447, + "loss": 1.1842, + "step": 416 + }, + { + "epoch": 1.7557894736842106, + "grad_norm": 0.51171875, + "learning_rate": 0.00012061641297358022, + "loss": 1.4594, + "step": 417 + }, + { + "epoch": 1.76, + "grad_norm": 0.34375, + "learning_rate": 0.00012028856840721974, + "loss": 1.1014, + "step": 418 + }, + { + "epoch": 1.7642105263157895, + "grad_norm": 0.640625, + "learning_rate": 0.00011996049625510901, + "loss": 0.7333, + "step": 419 + }, + { + "epoch": 1.768421052631579, + "grad_norm": 0.5859375, + "learning_rate": 0.00011963220019737691, + "loss": 1.2063, + "step": 420 + }, + { + "epoch": 1.768421052631579, + "eval_loss": 1.199410080909729, + "eval_runtime": 25.8809, + "eval_samples_per_second": 3.864, + "eval_steps_per_second": 1.932, + "step": 420 + }, + { + "epoch": 1.7726315789473683, + "grad_norm": 0.73046875, + "learning_rate": 0.00011930368391666398, + "loss": 1.2112, + "step": 421 + }, + { + "epoch": 1.776842105263158, + "grad_norm": 0.67578125, + "learning_rate": 0.00011897495109808107, + "loss": 1.0342, + "step": 422 + }, + { + "epoch": 1.7810526315789472, + "grad_norm": 0.376953125, + "learning_rate": 0.00011864600542916813, + "loss": 1.2239, + "step": 423 + }, + { + "epoch": 1.7852631578947369, + "grad_norm": 0.361328125, + "learning_rate": 0.00011831685059985262, + "loss": 0.9817, + "step": 424 + }, + { + "epoch": 1.7894736842105263, + "grad_norm": 0.30859375, + "learning_rate": 0.00011798749030240834, + "loss": 1.3617, + "step": 425 + }, + { + "epoch": 1.7936842105263158, + "grad_norm": 0.7265625, + "learning_rate": 0.00011765792823141384, + "loss": 1.2746, + "step": 426 + }, + { + "epoch": 1.7978947368421052, + "grad_norm": 0.484375, + "learning_rate": 0.00011732816808371115, + "loss": 1.471, + "step": 427 + }, + { + "epoch": 1.8021052631578947, + "grad_norm": 0.349609375, + "learning_rate": 0.00011699821355836409, + "loss": 1.1071, + "step": 428 + }, + { + "epoch": 1.8063157894736843, + "grad_norm": 0.26953125, + "learning_rate": 0.00011666806835661699, + "loss": 1.0702, + "step": 429 + }, + { + "epoch": 1.8105263157894735, + "grad_norm": 0.31640625, + "learning_rate": 0.00011633773618185302, + "loss": 1.2166, + "step": 430 + }, + { + "epoch": 1.8147368421052632, + "grad_norm": 0.36328125, + "learning_rate": 0.00011600722073955282, + "loss": 1.1926, + "step": 431 + }, + { + "epoch": 1.8189473684210526, + "grad_norm": 0.35546875, + "learning_rate": 0.00011567652573725262, + "loss": 1.0472, + "step": 432 + }, + { + "epoch": 1.823157894736842, + "grad_norm": 0.765625, + "learning_rate": 0.00011534565488450311, + "loss": 0.9258, + "step": 433 + }, + { + "epoch": 1.8273684210526315, + "grad_norm": 0.765625, + "learning_rate": 0.00011501461189282733, + "loss": 1.6986, + "step": 434 + }, + { + "epoch": 1.831578947368421, + "grad_norm": 0.392578125, + "learning_rate": 0.00011468340047567947, + "loss": 1.1726, + "step": 435 + }, + { + "epoch": 1.8357894736842106, + "grad_norm": 0.412109375, + "learning_rate": 0.00011435202434840287, + "loss": 1.2373, + "step": 436 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 0.6015625, + "learning_rate": 0.0001140204872281886, + "loss": 0.7262, + "step": 437 + }, + { + "epoch": 1.8442105263157895, + "grad_norm": 0.490234375, + "learning_rate": 0.0001136887928340336, + "loss": 1.0527, + "step": 438 + }, + { + "epoch": 1.848421052631579, + "grad_norm": 0.453125, + "learning_rate": 0.00011335694488669904, + "loss": 0.89, + "step": 439 + }, + { + "epoch": 1.8526315789473684, + "grad_norm": 0.55859375, + "learning_rate": 0.00011302494710866857, + "loss": 1.109, + "step": 440 + }, + { + "epoch": 1.8568421052631578, + "grad_norm": 0.5546875, + "learning_rate": 0.00011269280322410658, + "loss": 1.0818, + "step": 441 + }, + { + "epoch": 1.8610526315789473, + "grad_norm": 0.3671875, + "learning_rate": 0.00011236051695881633, + "loss": 1.2366, + "step": 442 + }, + { + "epoch": 1.865263157894737, + "grad_norm": 0.427734375, + "learning_rate": 0.0001120280920401983, + "loss": 1.1401, + "step": 443 + }, + { + "epoch": 1.8694736842105262, + "grad_norm": 0.5, + "learning_rate": 0.00011169553219720828, + "loss": 1.0061, + "step": 444 + }, + { + "epoch": 1.8736842105263158, + "grad_norm": 0.43359375, + "learning_rate": 0.0001113628411603155, + "loss": 1.2742, + "step": 445 + }, + { + "epoch": 1.8778947368421053, + "grad_norm": 0.375, + "learning_rate": 0.00011103002266146096, + "loss": 1.296, + "step": 446 + }, + { + "epoch": 1.8821052631578947, + "grad_norm": 0.7578125, + "learning_rate": 0.0001106970804340154, + "loss": 0.9062, + "step": 447 + }, + { + "epoch": 1.8863157894736842, + "grad_norm": 0.44140625, + "learning_rate": 0.0001103640182127375, + "loss": 0.85, + "step": 448 + }, + { + "epoch": 1.8905263157894736, + "grad_norm": 0.6171875, + "learning_rate": 0.00011003083973373187, + "loss": 1.0045, + "step": 449 + }, + { + "epoch": 1.8947368421052633, + "grad_norm": 0.50390625, + "learning_rate": 0.00010969754873440743, + "loss": 0.8616, + "step": 450 + }, + { + "epoch": 1.8989473684210525, + "grad_norm": 0.5546875, + "learning_rate": 0.0001093641489534351, + "loss": 0.9867, + "step": 451 + }, + { + "epoch": 1.9031578947368422, + "grad_norm": 0.462890625, + "learning_rate": 0.00010903064413070612, + "loss": 1.1836, + "step": 452 + }, + { + "epoch": 1.9073684210526316, + "grad_norm": 0.515625, + "learning_rate": 0.00010869703800729, + "loss": 0.9996, + "step": 453 + }, + { + "epoch": 1.911578947368421, + "grad_norm": 0.5078125, + "learning_rate": 0.00010836333432539272, + "loss": 1.4083, + "step": 454 + }, + { + "epoch": 1.9157894736842105, + "grad_norm": 0.474609375, + "learning_rate": 0.0001080295368283144, + "loss": 0.9889, + "step": 455 + }, + { + "epoch": 1.92, + "grad_norm": 0.30078125, + "learning_rate": 0.00010769564926040769, + "loss": 0.8322, + "step": 456 + }, + { + "epoch": 1.9242105263157896, + "grad_norm": 1.171875, + "learning_rate": 0.00010736167536703556, + "loss": 0.8793, + "step": 457 + }, + { + "epoch": 1.9284210526315788, + "grad_norm": 0.59765625, + "learning_rate": 0.0001070276188945293, + "loss": 1.1383, + "step": 458 + }, + { + "epoch": 1.9326315789473685, + "grad_norm": 0.625, + "learning_rate": 0.00010669348359014657, + "loss": 1.2608, + "step": 459 + }, + { + "epoch": 1.936842105263158, + "grad_norm": 0.49609375, + "learning_rate": 0.00010635927320202928, + "loss": 1.1444, + "step": 460 + }, + { + "epoch": 1.9410526315789474, + "grad_norm": 0.6796875, + "learning_rate": 0.00010602499147916159, + "loss": 1.1779, + "step": 461 + }, + { + "epoch": 1.9452631578947368, + "grad_norm": 0.390625, + "learning_rate": 0.00010569064217132791, + "loss": 1.2066, + "step": 462 + }, + { + "epoch": 1.9494736842105262, + "grad_norm": 0.32421875, + "learning_rate": 0.00010535622902907072, + "loss": 1.3065, + "step": 463 + }, + { + "epoch": 1.953684210526316, + "grad_norm": 0.45703125, + "learning_rate": 0.00010502175580364857, + "loss": 0.9167, + "step": 464 + }, + { + "epoch": 1.9578947368421051, + "grad_norm": 0.8203125, + "learning_rate": 0.00010468722624699401, + "loss": 1.059, + "step": 465 + }, + { + "epoch": 1.9621052631578948, + "grad_norm": 0.93359375, + "learning_rate": 0.00010435264411167148, + "loss": 0.8437, + "step": 466 + }, + { + "epoch": 1.9663157894736842, + "grad_norm": 0.4453125, + "learning_rate": 0.00010401801315083519, + "loss": 1.1105, + "step": 467 + }, + { + "epoch": 1.9705263157894737, + "grad_norm": 0.81640625, + "learning_rate": 0.0001036833371181871, + "loss": 1.5699, + "step": 468 + }, + { + "epoch": 1.9747368421052631, + "grad_norm": 0.359375, + "learning_rate": 0.00010334861976793465, + "loss": 1.0942, + "step": 469 + }, + { + "epoch": 1.9789473684210526, + "grad_norm": 0.4765625, + "learning_rate": 0.00010301386485474889, + "loss": 0.923, + "step": 470 + }, + { + "epoch": 1.9831578947368422, + "grad_norm": 0.498046875, + "learning_rate": 0.00010267907613372217, + "loss": 1.123, + "step": 471 + }, + { + "epoch": 1.9873684210526315, + "grad_norm": 0.45703125, + "learning_rate": 0.00010234425736032607, + "loss": 1.1807, + "step": 472 + }, + { + "epoch": 1.9915789473684211, + "grad_norm": 0.53515625, + "learning_rate": 0.00010200941229036926, + "loss": 1.2465, + "step": 473 + }, + { + "epoch": 1.9957894736842106, + "grad_norm": 0.3984375, + "learning_rate": 0.00010167454467995549, + "loss": 0.9394, + "step": 474 + } + ], + "logging_steps": 1, + "max_steps": 948, + "num_input_tokens_seen": 0, + "num_train_epochs": 4, + "save_steps": 237, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 7894666995302400.0, + "train_batch_size": 2, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-474/training_args.bin b/checkpoint-474/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..bd7f9976bbc503fbce44b538eec2dade4024c02a --- /dev/null +++ b/checkpoint-474/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee017c711219d88132625f3e50d1a67d1c6236a55d3f5a2ad9f499310d968cef +size 5944 diff --git a/checkpoint-711/README.md b/checkpoint-711/README.md new file mode 100644 index 0000000000000000000000000000000000000000..e1ccd431539a8f1507d8755a9c3ba5e5b2897978 --- /dev/null +++ b/checkpoint-711/README.md @@ -0,0 +1,202 @@ +--- +library_name: peft +base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **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] +### Framework versions + +- PEFT 0.11.1 \ No newline at end of file diff --git a/checkpoint-711/adapter_config.json b/checkpoint-711/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..cfa45a23cb958fe6ad106d4adf7f6169698c1fc8 --- /dev/null +++ b/checkpoint-711/adapter_config.json @@ -0,0 +1,34 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 16, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "v_proj", + "k_proj", + "q_proj", + "gate_proj", + "up_proj", + "down_proj", + "o_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-711/adapter_model.safetensors b/checkpoint-711/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8463a95cd1610315e507f4ec0cc529d151474100 --- /dev/null +++ b/checkpoint-711/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d357a366841f558231ed2389af28f3fe28086fe66e31a267d4f56d9156a9362b +size 50503848 diff --git a/checkpoint-711/optimizer.pt b/checkpoint-711/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..7ce99b0856f278eb139cfdd999793d56bd441913 --- /dev/null +++ b/checkpoint-711/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d0c973fd01cf5316029454305401c3a156d9fab6260ae3b9a4e66c840802f9f0 +size 202035770 diff --git a/checkpoint-711/rng_state.pth b/checkpoint-711/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..4b5393167336a4f734bedc797cdc13a2f60b00bf --- /dev/null +++ b/checkpoint-711/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:11d2e6162181dcdcfe51f923b36befadfc1a6aba4ef5f93190a0edf99f4ab509 +size 14244 diff --git a/checkpoint-711/scheduler.pt b/checkpoint-711/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..c633ab637dec626cd7824100334f633a5ee5d9a5 --- /dev/null +++ b/checkpoint-711/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da888c03aef9f24d8da0ae60bea1fbcb9a9e606b82616e3a258fa56bd3b1d022 +size 1064 diff --git a/checkpoint-711/special_tokens_map.json b/checkpoint-711/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1 --- /dev/null +++ b/checkpoint-711/special_tokens_map.json @@ -0,0 +1,24 @@ +{ + "bos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "", + "unk_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-711/tokenizer.model b/checkpoint-711/tokenizer.model new file mode 100644 index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899 --- /dev/null +++ b/checkpoint-711/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 +size 499723 diff --git a/checkpoint-711/tokenizer_config.json b/checkpoint-711/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..0773857a13ba5a27453a0b462624fe76e8e82a86 --- /dev/null +++ b/checkpoint-711/tokenizer_config.json @@ -0,0 +1,44 @@ +{ + "add_bos_token": true, + "add_eos_token": false, + "add_prefix_space": true, + "added_tokens_decoder": { + "0": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "1": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "2": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "", + "clean_up_tokenization_spaces": false, + "eos_token": "", + "legacy": false, + "model_max_length": 1000000000000000019884624838656, + "pad_token": "", + "padding_side": "right", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "LlamaTokenizer", + "unk_token": "", + "use_default_system_prompt": false, + "use_fast": true +} diff --git a/checkpoint-711/trainer_state.json b/checkpoint-711/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..7cb784f5633ae80356c6a837fab7fe26115c1818 --- /dev/null +++ b/checkpoint-711/trainer_state.json @@ -0,0 +1,5106 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.993684210526316, + "eval_steps": 60, + "global_step": 711, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.004210526315789474, + "grad_norm": 0.310546875, + "learning_rate": 2e-05, + "loss": 1.5605, + "step": 1 + }, + { + "epoch": 0.004210526315789474, + "eval_loss": 1.5264556407928467, + "eval_runtime": 26.3082, + "eval_samples_per_second": 3.801, + "eval_steps_per_second": 1.901, + "step": 1 + }, + { + "epoch": 0.008421052631578947, + "grad_norm": 0.306640625, + "learning_rate": 4e-05, + "loss": 1.4702, + "step": 2 + }, + { + "epoch": 0.01263157894736842, + "grad_norm": 0.2578125, + "learning_rate": 6e-05, + "loss": 1.4576, + "step": 3 + }, + { + "epoch": 0.016842105263157894, + "grad_norm": 0.65234375, + "learning_rate": 8e-05, + "loss": 1.5868, + "step": 4 + }, + { + "epoch": 0.021052631578947368, + "grad_norm": 0.56640625, + "learning_rate": 0.0001, + "loss": 1.5916, + "step": 5 + }, + { + "epoch": 0.02526315789473684, + "grad_norm": 0.5703125, + "learning_rate": 0.00012, + "loss": 1.7851, + "step": 6 + }, + { + "epoch": 0.029473684210526315, + "grad_norm": 0.3359375, + "learning_rate": 0.00014, + "loss": 1.5319, + "step": 7 + }, + { + "epoch": 0.03368421052631579, + "grad_norm": 0.39453125, + "learning_rate": 0.00016, + "loss": 1.8462, + "step": 8 + }, + { + "epoch": 0.037894736842105266, + "grad_norm": 0.484375, + "learning_rate": 0.00018, + "loss": 1.4219, + "step": 9 + }, + { + "epoch": 0.042105263157894736, + "grad_norm": 0.30859375, + "learning_rate": 0.0002, + "loss": 1.6555, + "step": 10 + }, + { + "epoch": 0.04631578947368421, + "grad_norm": 0.267578125, + "learning_rate": 0.0001999994391281196, + "loss": 1.1527, + "step": 11 + }, + { + "epoch": 0.05052631578947368, + "grad_norm": 0.298828125, + "learning_rate": 0.00019999775651876987, + "loss": 1.4265, + "step": 12 + }, + { + "epoch": 0.05473684210526316, + "grad_norm": 0.376953125, + "learning_rate": 0.00019999495219082545, + "loss": 1.5505, + "step": 13 + }, + { + "epoch": 0.05894736842105263, + "grad_norm": 0.380859375, + "learning_rate": 0.00019999102617574365, + "loss": 1.0313, + "step": 14 + }, + { + "epoch": 0.06315789473684211, + "grad_norm": 0.494140625, + "learning_rate": 0.0001999859785175643, + "loss": 1.439, + "step": 15 + }, + { + "epoch": 0.06736842105263158, + "grad_norm": 0.40234375, + "learning_rate": 0.00019997980927290927, + "loss": 1.3511, + "step": 16 + }, + { + "epoch": 0.07157894736842105, + "grad_norm": 0.390625, + "learning_rate": 0.0001999725185109816, + "loss": 1.2886, + "step": 17 + }, + { + "epoch": 0.07578947368421053, + "grad_norm": 0.388671875, + "learning_rate": 0.00019996410631356498, + "loss": 1.3057, + "step": 18 + }, + { + "epoch": 0.08, + "grad_norm": 0.6328125, + "learning_rate": 0.00019995457277502271, + "loss": 1.7375, + "step": 19 + }, + { + "epoch": 0.08421052631578947, + "grad_norm": 0.51953125, + "learning_rate": 0.00019994391800229666, + "loss": 1.3436, + "step": 20 + }, + { + "epoch": 0.08842105263157894, + "grad_norm": 0.345703125, + "learning_rate": 0.00019993214211490608, + "loss": 1.14, + "step": 21 + }, + { + "epoch": 0.09263157894736843, + "grad_norm": 0.357421875, + "learning_rate": 0.00019991924524494627, + "loss": 1.1322, + "step": 22 + }, + { + "epoch": 0.0968421052631579, + "grad_norm": 0.431640625, + "learning_rate": 0.00019990522753708705, + "loss": 1.4512, + "step": 23 + }, + { + "epoch": 0.10105263157894737, + "grad_norm": 0.2490234375, + "learning_rate": 0.00019989008914857116, + "loss": 1.3419, + "step": 24 + }, + { + "epoch": 0.10526315789473684, + "grad_norm": 0.78515625, + "learning_rate": 0.00019987383024921255, + "loss": 1.382, + "step": 25 + }, + { + "epoch": 0.10947368421052632, + "grad_norm": 0.56640625, + "learning_rate": 0.0001998564510213944, + "loss": 1.582, + "step": 26 + }, + { + "epoch": 0.11368421052631579, + "grad_norm": 0.5703125, + "learning_rate": 0.00019983795166006715, + "loss": 1.188, + "step": 27 + }, + { + "epoch": 0.11789473684210526, + "grad_norm": 0.46875, + "learning_rate": 0.00019981833237274618, + "loss": 1.4246, + "step": 28 + }, + { + "epoch": 0.12210526315789473, + "grad_norm": 0.3984375, + "learning_rate": 0.00019979759337950968, + "loss": 1.2828, + "step": 29 + }, + { + "epoch": 0.12631578947368421, + "grad_norm": 0.55078125, + "learning_rate": 0.00019977573491299598, + "loss": 1.6199, + "step": 30 + }, + { + "epoch": 0.13052631578947368, + "grad_norm": 0.8359375, + "learning_rate": 0.00019975275721840103, + "loss": 1.2398, + "step": 31 + }, + { + "epoch": 0.13473684210526315, + "grad_norm": 0.5, + "learning_rate": 0.00019972866055347572, + "loss": 1.4295, + "step": 32 + }, + { + "epoch": 0.13894736842105262, + "grad_norm": 0.5390625, + "learning_rate": 0.0001997034451885229, + "loss": 1.2814, + "step": 33 + }, + { + "epoch": 0.1431578947368421, + "grad_norm": 0.28515625, + "learning_rate": 0.0001996771114063943, + "loss": 1.532, + "step": 34 + }, + { + "epoch": 0.14736842105263157, + "grad_norm": 0.40625, + "learning_rate": 0.00019964965950248757, + "loss": 1.3516, + "step": 35 + }, + { + "epoch": 0.15157894736842106, + "grad_norm": 0.44921875, + "learning_rate": 0.00019962108978474263, + "loss": 1.1487, + "step": 36 + }, + { + "epoch": 0.15578947368421053, + "grad_norm": 0.310546875, + "learning_rate": 0.0001995914025736386, + "loss": 1.3543, + "step": 37 + }, + { + "epoch": 0.16, + "grad_norm": 0.306640625, + "learning_rate": 0.00019956059820218982, + "loss": 1.2986, + "step": 38 + }, + { + "epoch": 0.16421052631578947, + "grad_norm": 0.625, + "learning_rate": 0.00019952867701594247, + "loss": 1.0167, + "step": 39 + }, + { + "epoch": 0.16842105263157894, + "grad_norm": 0.306640625, + "learning_rate": 0.00019949563937297045, + "loss": 1.2287, + "step": 40 + }, + { + "epoch": 0.1726315789473684, + "grad_norm": 0.34765625, + "learning_rate": 0.00019946148564387145, + "loss": 1.0063, + "step": 41 + }, + { + "epoch": 0.17684210526315788, + "grad_norm": 0.94921875, + "learning_rate": 0.00019942621621176282, + "loss": 1.2712, + "step": 42 + }, + { + "epoch": 0.18105263157894738, + "grad_norm": 0.333984375, + "learning_rate": 0.00019938983147227716, + "loss": 1.2106, + "step": 43 + }, + { + "epoch": 0.18526315789473685, + "grad_norm": 0.6328125, + "learning_rate": 0.0001993523318335581, + "loss": 1.4887, + "step": 44 + }, + { + "epoch": 0.18947368421052632, + "grad_norm": 0.62890625, + "learning_rate": 0.00019931371771625544, + "loss": 1.2747, + "step": 45 + }, + { + "epoch": 0.1936842105263158, + "grad_norm": 0.2216796875, + "learning_rate": 0.00019927398955352061, + "loss": 1.1946, + "step": 46 + }, + { + "epoch": 0.19789473684210526, + "grad_norm": 0.54296875, + "learning_rate": 0.00019923314779100184, + "loss": 1.2467, + "step": 47 + }, + { + "epoch": 0.20210526315789473, + "grad_norm": 0.25390625, + "learning_rate": 0.00019919119288683908, + "loss": 1.5067, + "step": 48 + }, + { + "epoch": 0.2063157894736842, + "grad_norm": 0.404296875, + "learning_rate": 0.00019914812531165877, + "loss": 1.4291, + "step": 49 + }, + { + "epoch": 0.21052631578947367, + "grad_norm": 0.267578125, + "learning_rate": 0.00019910394554856876, + "loss": 1.0136, + "step": 50 + }, + { + "epoch": 0.21473684210526317, + "grad_norm": 1.328125, + "learning_rate": 0.00019905865409315288, + "loss": 1.2874, + "step": 51 + }, + { + "epoch": 0.21894736842105264, + "grad_norm": 0.244140625, + "learning_rate": 0.0001990122514534651, + "loss": 0.9588, + "step": 52 + }, + { + "epoch": 0.2231578947368421, + "grad_norm": 0.376953125, + "learning_rate": 0.0001989647381500242, + "loss": 1.3036, + "step": 53 + }, + { + "epoch": 0.22736842105263158, + "grad_norm": 0.5, + "learning_rate": 0.00019891611471580764, + "loss": 1.514, + "step": 54 + }, + { + "epoch": 0.23157894736842105, + "grad_norm": 0.423828125, + "learning_rate": 0.0001988663816962458, + "loss": 1.2444, + "step": 55 + }, + { + "epoch": 0.23578947368421052, + "grad_norm": 0.30078125, + "learning_rate": 0.00019881553964921572, + "loss": 1.2261, + "step": 56 + }, + { + "epoch": 0.24, + "grad_norm": 0.34765625, + "learning_rate": 0.00019876358914503492, + "loss": 1.2489, + "step": 57 + }, + { + "epoch": 0.24421052631578946, + "grad_norm": 0.431640625, + "learning_rate": 0.00019871053076645488, + "loss": 1.2134, + "step": 58 + }, + { + "epoch": 0.24842105263157896, + "grad_norm": 0.259765625, + "learning_rate": 0.00019865636510865467, + "loss": 1.1999, + "step": 59 + }, + { + "epoch": 0.25263157894736843, + "grad_norm": 0.2421875, + "learning_rate": 0.00019860109277923418, + "loss": 1.1485, + "step": 60 + }, + { + "epoch": 0.25263157894736843, + "eval_loss": 1.2386109828948975, + "eval_runtime": 25.8708, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.933, + "step": 60 + }, + { + "epoch": 0.25684210526315787, + "grad_norm": 0.5390625, + "learning_rate": 0.00019854471439820733, + "loss": 1.2963, + "step": 61 + }, + { + "epoch": 0.26105263157894737, + "grad_norm": 0.345703125, + "learning_rate": 0.00019848723059799506, + "loss": 1.097, + "step": 62 + }, + { + "epoch": 0.26526315789473687, + "grad_norm": 0.49609375, + "learning_rate": 0.00019842864202341835, + "loss": 1.1463, + "step": 63 + }, + { + "epoch": 0.2694736842105263, + "grad_norm": 0.267578125, + "learning_rate": 0.00019836894933169088, + "loss": 1.1921, + "step": 64 + }, + { + "epoch": 0.2736842105263158, + "grad_norm": 0.349609375, + "learning_rate": 0.00019830815319241168, + "loss": 1.1835, + "step": 65 + }, + { + "epoch": 0.27789473684210525, + "grad_norm": 0.333984375, + "learning_rate": 0.0001982462542875576, + "loss": 1.1998, + "step": 66 + }, + { + "epoch": 0.28210526315789475, + "grad_norm": 0.244140625, + "learning_rate": 0.00019818325331147585, + "loss": 1.1047, + "step": 67 + }, + { + "epoch": 0.2863157894736842, + "grad_norm": 0.244140625, + "learning_rate": 0.00019811915097087587, + "loss": 1.2452, + "step": 68 + }, + { + "epoch": 0.2905263157894737, + "grad_norm": 0.259765625, + "learning_rate": 0.00019805394798482167, + "loss": 1.1633, + "step": 69 + }, + { + "epoch": 0.29473684210526313, + "grad_norm": 0.44140625, + "learning_rate": 0.00019798764508472373, + "loss": 1.0081, + "step": 70 + }, + { + "epoch": 0.29894736842105263, + "grad_norm": 0.5703125, + "learning_rate": 0.00019792024301433064, + "loss": 0.9194, + "step": 71 + }, + { + "epoch": 0.3031578947368421, + "grad_norm": 0.357421875, + "learning_rate": 0.00019785174252972092, + "loss": 0.8091, + "step": 72 + }, + { + "epoch": 0.30736842105263157, + "grad_norm": 0.33984375, + "learning_rate": 0.00019778214439929452, + "loss": 1.1248, + "step": 73 + }, + { + "epoch": 0.31157894736842107, + "grad_norm": 0.23046875, + "learning_rate": 0.0001977114494037641, + "loss": 1.1204, + "step": 74 + }, + { + "epoch": 0.3157894736842105, + "grad_norm": 0.51171875, + "learning_rate": 0.0001976396583361464, + "loss": 1.1662, + "step": 75 + }, + { + "epoch": 0.32, + "grad_norm": 0.2578125, + "learning_rate": 0.00019756677200175315, + "loss": 1.1222, + "step": 76 + }, + { + "epoch": 0.32421052631578945, + "grad_norm": 0.796875, + "learning_rate": 0.00019749279121818235, + "loss": 1.0329, + "step": 77 + }, + { + "epoch": 0.32842105263157895, + "grad_norm": 0.2490234375, + "learning_rate": 0.0001974177168153088, + "loss": 1.1199, + "step": 78 + }, + { + "epoch": 0.33263157894736844, + "grad_norm": 0.26171875, + "learning_rate": 0.00019734154963527495, + "loss": 1.4391, + "step": 79 + }, + { + "epoch": 0.3368421052631579, + "grad_norm": 0.30078125, + "learning_rate": 0.0001972642905324813, + "loss": 1.3924, + "step": 80 + }, + { + "epoch": 0.3410526315789474, + "grad_norm": 0.208984375, + "learning_rate": 0.00019718594037357713, + "loss": 1.1518, + "step": 81 + }, + { + "epoch": 0.3452631578947368, + "grad_norm": 0.953125, + "learning_rate": 0.0001971065000374504, + "loss": 1.2606, + "step": 82 + }, + { + "epoch": 0.3494736842105263, + "grad_norm": 0.306640625, + "learning_rate": 0.0001970259704152181, + "loss": 1.2293, + "step": 83 + }, + { + "epoch": 0.35368421052631577, + "grad_norm": 0.6328125, + "learning_rate": 0.0001969443524102163, + "loss": 0.9564, + "step": 84 + }, + { + "epoch": 0.35789473684210527, + "grad_norm": 0.279296875, + "learning_rate": 0.00019686164693798986, + "loss": 1.2023, + "step": 85 + }, + { + "epoch": 0.36210526315789476, + "grad_norm": 0.263671875, + "learning_rate": 0.0001967778549262822, + "loss": 0.9322, + "step": 86 + }, + { + "epoch": 0.3663157894736842, + "grad_norm": 0.51171875, + "learning_rate": 0.00019669297731502507, + "loss": 1.2589, + "step": 87 + }, + { + "epoch": 0.3705263157894737, + "grad_norm": 0.392578125, + "learning_rate": 0.00019660701505632772, + "loss": 1.3188, + "step": 88 + }, + { + "epoch": 0.37473684210526315, + "grad_norm": 0.51953125, + "learning_rate": 0.00019651996911446647, + "loss": 1.0415, + "step": 89 + }, + { + "epoch": 0.37894736842105264, + "grad_norm": 0.427734375, + "learning_rate": 0.0001964318404658737, + "loss": 1.467, + "step": 90 + }, + { + "epoch": 0.3831578947368421, + "grad_norm": 0.4453125, + "learning_rate": 0.00019634263009912705, + "loss": 1.5527, + "step": 91 + }, + { + "epoch": 0.3873684210526316, + "grad_norm": 0.30859375, + "learning_rate": 0.00019625233901493822, + "loss": 1.2043, + "step": 92 + }, + { + "epoch": 0.391578947368421, + "grad_norm": 0.279296875, + "learning_rate": 0.00019616096822614186, + "loss": 0.9677, + "step": 93 + }, + { + "epoch": 0.3957894736842105, + "grad_norm": 0.46484375, + "learning_rate": 0.000196068518757684, + "loss": 1.1942, + "step": 94 + }, + { + "epoch": 0.4, + "grad_norm": 0.435546875, + "learning_rate": 0.00019597499164661093, + "loss": 1.3481, + "step": 95 + }, + { + "epoch": 0.40421052631578946, + "grad_norm": 0.291015625, + "learning_rate": 0.00019588038794205703, + "loss": 1.2776, + "step": 96 + }, + { + "epoch": 0.40842105263157896, + "grad_norm": 0.3828125, + "learning_rate": 0.00019578470870523359, + "loss": 1.4808, + "step": 97 + }, + { + "epoch": 0.4126315789473684, + "grad_norm": 0.28125, + "learning_rate": 0.00019568795500941635, + "loss": 1.3385, + "step": 98 + }, + { + "epoch": 0.4168421052631579, + "grad_norm": 0.484375, + "learning_rate": 0.00019559012793993392, + "loss": 1.5787, + "step": 99 + }, + { + "epoch": 0.42105263157894735, + "grad_norm": 0.298828125, + "learning_rate": 0.00019549122859415538, + "loss": 1.2151, + "step": 100 + }, + { + "epoch": 0.42526315789473684, + "grad_norm": 0.65234375, + "learning_rate": 0.0001953912580814779, + "loss": 1.0332, + "step": 101 + }, + { + "epoch": 0.42947368421052634, + "grad_norm": 0.37890625, + "learning_rate": 0.00019529021752331453, + "loss": 1.2677, + "step": 102 + }, + { + "epoch": 0.4336842105263158, + "grad_norm": 0.265625, + "learning_rate": 0.0001951881080530814, + "loss": 1.0672, + "step": 103 + }, + { + "epoch": 0.4378947368421053, + "grad_norm": 0.498046875, + "learning_rate": 0.00019508493081618513, + "loss": 1.1736, + "step": 104 + }, + { + "epoch": 0.4421052631578947, + "grad_norm": 0.298828125, + "learning_rate": 0.00019498068697000994, + "loss": 1.1174, + "step": 105 + }, + { + "epoch": 0.4463157894736842, + "grad_norm": 0.169921875, + "learning_rate": 0.00019487537768390464, + "loss": 1.0712, + "step": 106 + }, + { + "epoch": 0.45052631578947366, + "grad_norm": 0.55078125, + "learning_rate": 0.0001947690041391696, + "loss": 1.4496, + "step": 107 + }, + { + "epoch": 0.45473684210526316, + "grad_norm": 0.47265625, + "learning_rate": 0.00019466156752904343, + "loss": 1.2413, + "step": 108 + }, + { + "epoch": 0.4589473684210526, + "grad_norm": 0.6015625, + "learning_rate": 0.0001945530690586896, + "loss": 1.0208, + "step": 109 + }, + { + "epoch": 0.4631578947368421, + "grad_norm": 0.35546875, + "learning_rate": 0.0001944435099451829, + "loss": 1.3997, + "step": 110 + }, + { + "epoch": 0.4673684210526316, + "grad_norm": 0.330078125, + "learning_rate": 0.00019433289141749586, + "loss": 1.2914, + "step": 111 + }, + { + "epoch": 0.47157894736842104, + "grad_norm": 0.63671875, + "learning_rate": 0.00019422121471648497, + "loss": 0.8134, + "step": 112 + }, + { + "epoch": 0.47578947368421054, + "grad_norm": 0.30859375, + "learning_rate": 0.00019410848109487658, + "loss": 1.1282, + "step": 113 + }, + { + "epoch": 0.48, + "grad_norm": 0.5625, + "learning_rate": 0.0001939946918172531, + "loss": 1.3254, + "step": 114 + }, + { + "epoch": 0.4842105263157895, + "grad_norm": 0.30859375, + "learning_rate": 0.00019387984816003867, + "loss": 1.6428, + "step": 115 + }, + { + "epoch": 0.4884210526315789, + "grad_norm": 0.33203125, + "learning_rate": 0.00019376395141148476, + "loss": 1.0688, + "step": 116 + }, + { + "epoch": 0.4926315789473684, + "grad_norm": 0.6953125, + "learning_rate": 0.00019364700287165604, + "loss": 1.0969, + "step": 117 + }, + { + "epoch": 0.4968421052631579, + "grad_norm": 1.296875, + "learning_rate": 0.00019352900385241536, + "loss": 1.5249, + "step": 118 + }, + { + "epoch": 0.5010526315789474, + "grad_norm": 0.2001953125, + "learning_rate": 0.00019340995567740938, + "loss": 1.2095, + "step": 119 + }, + { + "epoch": 0.5052631578947369, + "grad_norm": 0.53515625, + "learning_rate": 0.0001932898596820536, + "loss": 1.1249, + "step": 120 + }, + { + "epoch": 0.5052631578947369, + "eval_loss": 1.2166697978973389, + "eval_runtime": 25.8713, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.933, + "step": 120 + }, + { + "epoch": 0.5094736842105263, + "grad_norm": 0.423828125, + "learning_rate": 0.00019316871721351733, + "loss": 1.5017, + "step": 121 + }, + { + "epoch": 0.5136842105263157, + "grad_norm": 0.7421875, + "learning_rate": 0.0001930465296307087, + "loss": 1.0994, + "step": 122 + }, + { + "epoch": 0.5178947368421053, + "grad_norm": 0.21875, + "learning_rate": 0.0001929232983042592, + "loss": 1.0339, + "step": 123 + }, + { + "epoch": 0.5221052631578947, + "grad_norm": 0.6328125, + "learning_rate": 0.00019279902461650866, + "loss": 1.1317, + "step": 124 + }, + { + "epoch": 0.5263157894736842, + "grad_norm": 0.3984375, + "learning_rate": 0.00019267370996148934, + "loss": 0.9421, + "step": 125 + }, + { + "epoch": 0.5305263157894737, + "grad_norm": 0.396484375, + "learning_rate": 0.00019254735574491058, + "loss": 1.3029, + "step": 126 + }, + { + "epoch": 0.5347368421052632, + "grad_norm": 0.8359375, + "learning_rate": 0.00019241996338414296, + "loss": 1.3548, + "step": 127 + }, + { + "epoch": 0.5389473684210526, + "grad_norm": 0.84765625, + "learning_rate": 0.00019229153430820232, + "loss": 1.0865, + "step": 128 + }, + { + "epoch": 0.5431578947368421, + "grad_norm": 0.453125, + "learning_rate": 0.00019216206995773373, + "loss": 0.8151, + "step": 129 + }, + { + "epoch": 0.5473684210526316, + "grad_norm": 0.8046875, + "learning_rate": 0.0001920315717849956, + "loss": 1.3576, + "step": 130 + }, + { + "epoch": 0.5515789473684211, + "grad_norm": 0.82421875, + "learning_rate": 0.00019190004125384293, + "loss": 1.2439, + "step": 131 + }, + { + "epoch": 0.5557894736842105, + "grad_norm": 0.56640625, + "learning_rate": 0.0001917674798397113, + "loss": 0.9743, + "step": 132 + }, + { + "epoch": 0.56, + "grad_norm": 1.1015625, + "learning_rate": 0.0001916338890296001, + "loss": 1.3315, + "step": 133 + }, + { + "epoch": 0.5642105263157895, + "grad_norm": 0.333984375, + "learning_rate": 0.00019149927032205587, + "loss": 1.3268, + "step": 134 + }, + { + "epoch": 0.5684210526315789, + "grad_norm": 0.388671875, + "learning_rate": 0.00019136362522715562, + "loss": 1.4751, + "step": 135 + }, + { + "epoch": 0.5726315789473684, + "grad_norm": 0.2734375, + "learning_rate": 0.00019122695526648968, + "loss": 1.1219, + "step": 136 + }, + { + "epoch": 0.5768421052631579, + "grad_norm": 1.0703125, + "learning_rate": 0.00019108926197314486, + "loss": 1.4059, + "step": 137 + }, + { + "epoch": 0.5810526315789474, + "grad_norm": 0.33203125, + "learning_rate": 0.00019095054689168705, + "loss": 1.3282, + "step": 138 + }, + { + "epoch": 0.5852631578947368, + "grad_norm": 0.6171875, + "learning_rate": 0.00019081081157814404, + "loss": 1.6334, + "step": 139 + }, + { + "epoch": 0.5894736842105263, + "grad_norm": 0.359375, + "learning_rate": 0.00019067005759998797, + "loss": 1.159, + "step": 140 + }, + { + "epoch": 0.5936842105263158, + "grad_norm": 0.470703125, + "learning_rate": 0.00019052828653611786, + "loss": 1.3261, + "step": 141 + }, + { + "epoch": 0.5978947368421053, + "grad_norm": 0.26171875, + "learning_rate": 0.0001903854999768417, + "loss": 1.2041, + "step": 142 + }, + { + "epoch": 0.6021052631578947, + "grad_norm": 0.27734375, + "learning_rate": 0.00019024169952385885, + "loss": 1.1998, + "step": 143 + }, + { + "epoch": 0.6063157894736843, + "grad_norm": 0.56640625, + "learning_rate": 0.0001900968867902419, + "loss": 1.6427, + "step": 144 + }, + { + "epoch": 0.6105263157894737, + "grad_norm": 0.419921875, + "learning_rate": 0.00018995106340041873, + "loss": 1.1968, + "step": 145 + }, + { + "epoch": 0.6147368421052631, + "grad_norm": 0.671875, + "learning_rate": 0.00018980423099015402, + "loss": 0.8641, + "step": 146 + }, + { + "epoch": 0.6189473684210526, + "grad_norm": 0.333984375, + "learning_rate": 0.00018965639120653121, + "loss": 1.4518, + "step": 147 + }, + { + "epoch": 0.6231578947368421, + "grad_norm": 0.423828125, + "learning_rate": 0.00018950754570793384, + "loss": 1.3077, + "step": 148 + }, + { + "epoch": 0.6273684210526316, + "grad_norm": 0.85546875, + "learning_rate": 0.00018935769616402703, + "loss": 1.2077, + "step": 149 + }, + { + "epoch": 0.631578947368421, + "grad_norm": 0.224609375, + "learning_rate": 0.00018920684425573865, + "loss": 1.0868, + "step": 150 + }, + { + "epoch": 0.6357894736842106, + "grad_norm": 0.25390625, + "learning_rate": 0.00018905499167524062, + "loss": 1.2018, + "step": 151 + }, + { + "epoch": 0.64, + "grad_norm": 0.498046875, + "learning_rate": 0.00018890214012592975, + "loss": 1.1675, + "step": 152 + }, + { + "epoch": 0.6442105263157895, + "grad_norm": 0.2451171875, + "learning_rate": 0.00018874829132240874, + "loss": 1.2348, + "step": 153 + }, + { + "epoch": 0.6484210526315789, + "grad_norm": 0.33984375, + "learning_rate": 0.000188593446990467, + "loss": 1.3202, + "step": 154 + }, + { + "epoch": 0.6526315789473685, + "grad_norm": 1.0078125, + "learning_rate": 0.0001884376088670611, + "loss": 1.0171, + "step": 155 + }, + { + "epoch": 0.6568421052631579, + "grad_norm": 0.318359375, + "learning_rate": 0.00018828077870029552, + "loss": 1.219, + "step": 156 + }, + { + "epoch": 0.6610526315789473, + "grad_norm": 0.369140625, + "learning_rate": 0.00018812295824940285, + "loss": 0.8956, + "step": 157 + }, + { + "epoch": 0.6652631578947369, + "grad_norm": 0.328125, + "learning_rate": 0.00018796414928472417, + "loss": 0.9322, + "step": 158 + }, + { + "epoch": 0.6694736842105263, + "grad_norm": 0.2890625, + "learning_rate": 0.0001878043535876891, + "loss": 1.2062, + "step": 159 + }, + { + "epoch": 0.6736842105263158, + "grad_norm": 0.41015625, + "learning_rate": 0.0001876435729507959, + "loss": 1.3113, + "step": 160 + }, + { + "epoch": 0.6778947368421052, + "grad_norm": 0.53515625, + "learning_rate": 0.00018748180917759136, + "loss": 1.1539, + "step": 161 + }, + { + "epoch": 0.6821052631578948, + "grad_norm": 0.279296875, + "learning_rate": 0.0001873190640826505, + "loss": 1.3287, + "step": 162 + }, + { + "epoch": 0.6863157894736842, + "grad_norm": 0.298828125, + "learning_rate": 0.00018715533949155625, + "loss": 1.0616, + "step": 163 + }, + { + "epoch": 0.6905263157894737, + "grad_norm": 0.388671875, + "learning_rate": 0.00018699063724087904, + "loss": 1.0171, + "step": 164 + }, + { + "epoch": 0.6947368421052632, + "grad_norm": 0.32421875, + "learning_rate": 0.00018682495917815608, + "loss": 1.1113, + "step": 165 + }, + { + "epoch": 0.6989473684210527, + "grad_norm": 0.46875, + "learning_rate": 0.00018665830716187065, + "loss": 0.8351, + "step": 166 + }, + { + "epoch": 0.7031578947368421, + "grad_norm": 0.6015625, + "learning_rate": 0.0001864906830614314, + "loss": 1.2181, + "step": 167 + }, + { + "epoch": 0.7073684210526315, + "grad_norm": 0.71875, + "learning_rate": 0.0001863220887571512, + "loss": 1.1843, + "step": 168 + }, + { + "epoch": 0.7115789473684211, + "grad_norm": 0.51953125, + "learning_rate": 0.00018615252614022617, + "loss": 1.3809, + "step": 169 + }, + { + "epoch": 0.7157894736842105, + "grad_norm": 0.34375, + "learning_rate": 0.0001859819971127143, + "loss": 1.4288, + "step": 170 + }, + { + "epoch": 0.72, + "grad_norm": 0.41796875, + "learning_rate": 0.00018581050358751445, + "loss": 1.1449, + "step": 171 + }, + { + "epoch": 0.7242105263157895, + "grad_norm": 0.578125, + "learning_rate": 0.00018563804748834438, + "loss": 1.0703, + "step": 172 + }, + { + "epoch": 0.728421052631579, + "grad_norm": 0.43359375, + "learning_rate": 0.00018546463074971977, + "loss": 1.2093, + "step": 173 + }, + { + "epoch": 0.7326315789473684, + "grad_norm": 0.26171875, + "learning_rate": 0.000185290255316932, + "loss": 0.7346, + "step": 174 + }, + { + "epoch": 0.7368421052631579, + "grad_norm": 0.369140625, + "learning_rate": 0.00018511492314602656, + "loss": 1.2782, + "step": 175 + }, + { + "epoch": 0.7410526315789474, + "grad_norm": 0.56640625, + "learning_rate": 0.00018493863620378122, + "loss": 1.0844, + "step": 176 + }, + { + "epoch": 0.7452631578947368, + "grad_norm": 0.2470703125, + "learning_rate": 0.00018476139646768372, + "loss": 1.1771, + "step": 177 + }, + { + "epoch": 0.7494736842105263, + "grad_norm": 0.228515625, + "learning_rate": 0.00018458320592590975, + "loss": 1.2, + "step": 178 + }, + { + "epoch": 0.7536842105263157, + "grad_norm": 0.310546875, + "learning_rate": 0.00018440406657730058, + "loss": 1.2248, + "step": 179 + }, + { + "epoch": 0.7578947368421053, + "grad_norm": 0.263671875, + "learning_rate": 0.00018422398043134067, + "loss": 1.3675, + "step": 180 + }, + { + "epoch": 0.7578947368421053, + "eval_loss": 1.2129631042480469, + "eval_runtime": 25.9253, + "eval_samples_per_second": 3.857, + "eval_steps_per_second": 1.929, + "step": 180 + }, + { + "epoch": 0.7621052631578947, + "grad_norm": 0.2412109375, + "learning_rate": 0.0001840429495081352, + "loss": 1.2456, + "step": 181 + }, + { + "epoch": 0.7663157894736842, + "grad_norm": 0.25390625, + "learning_rate": 0.00018386097583838714, + "loss": 1.2617, + "step": 182 + }, + { + "epoch": 0.7705263157894737, + "grad_norm": 0.283203125, + "learning_rate": 0.00018367806146337483, + "loss": 1.2708, + "step": 183 + }, + { + "epoch": 0.7747368421052632, + "grad_norm": 0.318359375, + "learning_rate": 0.00018349420843492888, + "loss": 1.3771, + "step": 184 + }, + { + "epoch": 0.7789473684210526, + "grad_norm": 0.73046875, + "learning_rate": 0.00018330941881540915, + "loss": 1.3123, + "step": 185 + }, + { + "epoch": 0.783157894736842, + "grad_norm": 0.46875, + "learning_rate": 0.00018312369467768166, + "loss": 1.1952, + "step": 186 + }, + { + "epoch": 0.7873684210526316, + "grad_norm": 0.7578125, + "learning_rate": 0.00018293703810509537, + "loss": 1.426, + "step": 187 + }, + { + "epoch": 0.791578947368421, + "grad_norm": 0.33203125, + "learning_rate": 0.0001827494511914587, + "loss": 1.3205, + "step": 188 + }, + { + "epoch": 0.7957894736842105, + "grad_norm": 0.365234375, + "learning_rate": 0.00018256093604101617, + "loss": 1.2221, + "step": 189 + }, + { + "epoch": 0.8, + "grad_norm": 0.36328125, + "learning_rate": 0.0001823714947684247, + "loss": 1.4212, + "step": 190 + }, + { + "epoch": 0.8042105263157895, + "grad_norm": 0.349609375, + "learning_rate": 0.00018218112949872995, + "loss": 1.2183, + "step": 191 + }, + { + "epoch": 0.8084210526315789, + "grad_norm": 0.392578125, + "learning_rate": 0.00018198984236734246, + "loss": 0.9922, + "step": 192 + }, + { + "epoch": 0.8126315789473684, + "grad_norm": 0.373046875, + "learning_rate": 0.0001817976355200137, + "loss": 1.334, + "step": 193 + }, + { + "epoch": 0.8168421052631579, + "grad_norm": 0.30859375, + "learning_rate": 0.000181604511112812, + "loss": 1.1052, + "step": 194 + }, + { + "epoch": 0.8210526315789474, + "grad_norm": 0.349609375, + "learning_rate": 0.0001814104713120983, + "loss": 0.8112, + "step": 195 + }, + { + "epoch": 0.8252631578947368, + "grad_norm": 0.35546875, + "learning_rate": 0.000181215518294502, + "loss": 1.2507, + "step": 196 + }, + { + "epoch": 0.8294736842105264, + "grad_norm": 0.671875, + "learning_rate": 0.0001810196542468964, + "loss": 1.1228, + "step": 197 + }, + { + "epoch": 0.8336842105263158, + "grad_norm": 0.41015625, + "learning_rate": 0.00018082288136637422, + "loss": 1.3562, + "step": 198 + }, + { + "epoch": 0.8378947368421052, + "grad_norm": 0.302734375, + "learning_rate": 0.000180625201860223, + "loss": 0.7766, + "step": 199 + }, + { + "epoch": 0.8421052631578947, + "grad_norm": 0.322265625, + "learning_rate": 0.00018042661794590023, + "loss": 1.2885, + "step": 200 + }, + { + "epoch": 0.8463157894736842, + "grad_norm": 0.609375, + "learning_rate": 0.0001802271318510086, + "loss": 1.271, + "step": 201 + }, + { + "epoch": 0.8505263157894737, + "grad_norm": 0.341796875, + "learning_rate": 0.00018002674581327094, + "loss": 1.3777, + "step": 202 + }, + { + "epoch": 0.8547368421052631, + "grad_norm": 0.3828125, + "learning_rate": 0.00017982546208050515, + "loss": 1.32, + "step": 203 + }, + { + "epoch": 0.8589473684210527, + "grad_norm": 0.392578125, + "learning_rate": 0.00017962328291059888, + "loss": 1.1265, + "step": 204 + }, + { + "epoch": 0.8631578947368421, + "grad_norm": 0.73046875, + "learning_rate": 0.00017942021057148436, + "loss": 1.523, + "step": 205 + }, + { + "epoch": 0.8673684210526316, + "grad_norm": 1.046875, + "learning_rate": 0.00017921624734111292, + "loss": 1.0251, + "step": 206 + }, + { + "epoch": 0.871578947368421, + "grad_norm": 0.490234375, + "learning_rate": 0.0001790113955074294, + "loss": 1.2665, + "step": 207 + }, + { + "epoch": 0.8757894736842106, + "grad_norm": 0.6875, + "learning_rate": 0.0001788056573683464, + "loss": 1.2244, + "step": 208 + }, + { + "epoch": 0.88, + "grad_norm": 0.2177734375, + "learning_rate": 0.0001785990352317187, + "loss": 1.1621, + "step": 209 + }, + { + "epoch": 0.8842105263157894, + "grad_norm": 0.4921875, + "learning_rate": 0.00017839153141531718, + "loss": 1.3182, + "step": 210 + }, + { + "epoch": 0.888421052631579, + "grad_norm": 0.388671875, + "learning_rate": 0.000178183148246803, + "loss": 1.4508, + "step": 211 + }, + { + "epoch": 0.8926315789473684, + "grad_norm": 0.388671875, + "learning_rate": 0.00017797388806370132, + "loss": 1.1155, + "step": 212 + }, + { + "epoch": 0.8968421052631579, + "grad_norm": 0.416015625, + "learning_rate": 0.00017776375321337521, + "loss": 0.9191, + "step": 213 + }, + { + "epoch": 0.9010526315789473, + "grad_norm": 0.2431640625, + "learning_rate": 0.00017755274605299923, + "loss": 1.1577, + "step": 214 + }, + { + "epoch": 0.9052631578947369, + "grad_norm": 0.412109375, + "learning_rate": 0.00017734086894953305, + "loss": 1.372, + "step": 215 + }, + { + "epoch": 0.9094736842105263, + "grad_norm": 0.353515625, + "learning_rate": 0.00017712812427969485, + "loss": 1.1932, + "step": 216 + }, + { + "epoch": 0.9136842105263158, + "grad_norm": 0.291015625, + "learning_rate": 0.0001769145144299347, + "loss": 1.3167, + "step": 217 + }, + { + "epoch": 0.9178947368421052, + "grad_norm": 0.255859375, + "learning_rate": 0.00017670004179640774, + "loss": 1.2357, + "step": 218 + }, + { + "epoch": 0.9221052631578948, + "grad_norm": 0.5703125, + "learning_rate": 0.00017648470878494737, + "loss": 0.9556, + "step": 219 + }, + { + "epoch": 0.9263157894736842, + "grad_norm": 0.36328125, + "learning_rate": 0.0001762685178110382, + "loss": 1.2047, + "step": 220 + }, + { + "epoch": 0.9305263157894736, + "grad_norm": 0.271484375, + "learning_rate": 0.00017605147129978903, + "loss": 1.1935, + "step": 221 + }, + { + "epoch": 0.9347368421052632, + "grad_norm": 0.400390625, + "learning_rate": 0.0001758335716859055, + "loss": 1.3426, + "step": 222 + }, + { + "epoch": 0.9389473684210526, + "grad_norm": 0.73046875, + "learning_rate": 0.000175614821413663, + "loss": 1.5181, + "step": 223 + }, + { + "epoch": 0.9431578947368421, + "grad_norm": 0.3203125, + "learning_rate": 0.00017539522293687898, + "loss": 1.0962, + "step": 224 + }, + { + "epoch": 0.9473684210526315, + "grad_norm": 0.232421875, + "learning_rate": 0.00017517477871888568, + "loss": 1.3922, + "step": 225 + }, + { + "epoch": 0.9515789473684211, + "grad_norm": 0.5546875, + "learning_rate": 0.00017495349123250242, + "loss": 1.2535, + "step": 226 + }, + { + "epoch": 0.9557894736842105, + "grad_norm": 0.296875, + "learning_rate": 0.00017473136296000772, + "loss": 1.1987, + "step": 227 + }, + { + "epoch": 0.96, + "grad_norm": 0.77734375, + "learning_rate": 0.00017450839639311162, + "loss": 1.4301, + "step": 228 + }, + { + "epoch": 0.9642105263157895, + "grad_norm": 0.35546875, + "learning_rate": 0.00017428459403292767, + "loss": 1.3164, + "step": 229 + }, + { + "epoch": 0.968421052631579, + "grad_norm": 0.419921875, + "learning_rate": 0.00017405995838994494, + "loss": 1.1554, + "step": 230 + }, + { + "epoch": 0.9726315789473684, + "grad_norm": 0.3125, + "learning_rate": 0.00017383449198399964, + "loss": 1.3036, + "step": 231 + }, + { + "epoch": 0.9768421052631578, + "grad_norm": 0.267578125, + "learning_rate": 0.00017360819734424715, + "loss": 1.2797, + "step": 232 + }, + { + "epoch": 0.9810526315789474, + "grad_norm": 0.412109375, + "learning_rate": 0.00017338107700913354, + "loss": 1.1202, + "step": 233 + }, + { + "epoch": 0.9852631578947368, + "grad_norm": 0.365234375, + "learning_rate": 0.0001731531335263669, + "loss": 0.8187, + "step": 234 + }, + { + "epoch": 0.9894736842105263, + "grad_norm": 0.5546875, + "learning_rate": 0.0001729243694528891, + "loss": 1.1702, + "step": 235 + }, + { + "epoch": 0.9936842105263158, + "grad_norm": 0.5546875, + "learning_rate": 0.00017269478735484683, + "loss": 1.1406, + "step": 236 + }, + { + "epoch": 0.9978947368421053, + "grad_norm": 0.3359375, + "learning_rate": 0.00017246438980756293, + "loss": 0.9633, + "step": 237 + }, + { + "epoch": 1.0021052631578948, + "grad_norm": 0.359375, + "learning_rate": 0.00017223317939550753, + "loss": 1.2181, + "step": 238 + }, + { + "epoch": 1.0063157894736843, + "grad_norm": 0.2138671875, + "learning_rate": 0.00017200115871226904, + "loss": 1.2035, + "step": 239 + }, + { + "epoch": 1.0105263157894737, + "grad_norm": 0.3203125, + "learning_rate": 0.00017176833036052495, + "loss": 1.3449, + "step": 240 + }, + { + "epoch": 1.0105263157894737, + "eval_loss": 1.1994891166687012, + "eval_runtime": 25.9787, + "eval_samples_per_second": 3.849, + "eval_steps_per_second": 1.925, + "step": 240 + }, + { + "epoch": 1.0147368421052632, + "grad_norm": 0.375, + "learning_rate": 0.00017153469695201277, + "loss": 1.0695, + "step": 241 + }, + { + "epoch": 1.0189473684210526, + "grad_norm": 0.578125, + "learning_rate": 0.0001713002611075007, + "loss": 0.8898, + "step": 242 + }, + { + "epoch": 1.023157894736842, + "grad_norm": 0.203125, + "learning_rate": 0.0001710650254567582, + "loss": 1.087, + "step": 243 + }, + { + "epoch": 1.0273684210526315, + "grad_norm": 0.3828125, + "learning_rate": 0.0001708289926385265, + "loss": 1.1739, + "step": 244 + }, + { + "epoch": 1.0315789473684212, + "grad_norm": 0.3515625, + "learning_rate": 0.000170592165300489, + "loss": 1.0413, + "step": 245 + }, + { + "epoch": 1.0357894736842106, + "grad_norm": 0.376953125, + "learning_rate": 0.0001703545460992416, + "loss": 1.0745, + "step": 246 + }, + { + "epoch": 1.04, + "grad_norm": 0.25, + "learning_rate": 0.00017011613770026292, + "loss": 1.2494, + "step": 247 + }, + { + "epoch": 1.0442105263157895, + "grad_norm": 0.419921875, + "learning_rate": 0.00016987694277788417, + "loss": 0.7568, + "step": 248 + }, + { + "epoch": 1.048421052631579, + "grad_norm": 0.6015625, + "learning_rate": 0.0001696369640152596, + "loss": 1.2675, + "step": 249 + }, + { + "epoch": 1.0526315789473684, + "grad_norm": 0.67578125, + "learning_rate": 0.0001693962041043359, + "loss": 1.0456, + "step": 250 + }, + { + "epoch": 1.0568421052631578, + "grad_norm": 0.5234375, + "learning_rate": 0.00016915466574582246, + "loss": 1.1145, + "step": 251 + }, + { + "epoch": 1.0610526315789475, + "grad_norm": 0.279296875, + "learning_rate": 0.00016891235164916065, + "loss": 1.1659, + "step": 252 + }, + { + "epoch": 1.065263157894737, + "grad_norm": 0.369140625, + "learning_rate": 0.0001686692645324938, + "loss": 0.9993, + "step": 253 + }, + { + "epoch": 1.0694736842105264, + "grad_norm": 0.78125, + "learning_rate": 0.00016842540712263637, + "loss": 1.2847, + "step": 254 + }, + { + "epoch": 1.0736842105263158, + "grad_norm": 0.48046875, + "learning_rate": 0.0001681807821550438, + "loss": 1.1483, + "step": 255 + }, + { + "epoch": 1.0778947368421052, + "grad_norm": 0.73046875, + "learning_rate": 0.00016793539237378128, + "loss": 0.7895, + "step": 256 + }, + { + "epoch": 1.0821052631578947, + "grad_norm": 0.30859375, + "learning_rate": 0.00016768924053149338, + "loss": 1.2321, + "step": 257 + }, + { + "epoch": 1.0863157894736841, + "grad_norm": 0.291015625, + "learning_rate": 0.00016744232938937308, + "loss": 1.0098, + "step": 258 + }, + { + "epoch": 1.0905263157894738, + "grad_norm": 0.470703125, + "learning_rate": 0.00016719466171713073, + "loss": 1.3362, + "step": 259 + }, + { + "epoch": 1.0947368421052632, + "grad_norm": 0.2890625, + "learning_rate": 0.0001669462402929629, + "loss": 1.2487, + "step": 260 + }, + { + "epoch": 1.0989473684210527, + "grad_norm": 0.39453125, + "learning_rate": 0.00016669706790352148, + "loss": 1.1732, + "step": 261 + }, + { + "epoch": 1.1031578947368421, + "grad_norm": 0.28125, + "learning_rate": 0.00016644714734388217, + "loss": 1.1493, + "step": 262 + }, + { + "epoch": 1.1073684210526316, + "grad_norm": 0.314453125, + "learning_rate": 0.0001661964814175133, + "loss": 1.0521, + "step": 263 + }, + { + "epoch": 1.111578947368421, + "grad_norm": 0.53125, + "learning_rate": 0.00016594507293624425, + "loss": 1.07, + "step": 264 + }, + { + "epoch": 1.1157894736842104, + "grad_norm": 0.2890625, + "learning_rate": 0.00016569292472023392, + "loss": 1.1393, + "step": 265 + }, + { + "epoch": 1.12, + "grad_norm": 0.390625, + "learning_rate": 0.00016544003959793925, + "loss": 1.2059, + "step": 266 + }, + { + "epoch": 1.1242105263157895, + "grad_norm": 0.494140625, + "learning_rate": 0.0001651864204060833, + "loss": 0.9832, + "step": 267 + }, + { + "epoch": 1.128421052631579, + "grad_norm": 0.376953125, + "learning_rate": 0.00016493206998962354, + "loss": 0.9154, + "step": 268 + }, + { + "epoch": 1.1326315789473684, + "grad_norm": 0.39453125, + "learning_rate": 0.00016467699120171987, + "loss": 1.0839, + "step": 269 + }, + { + "epoch": 1.1368421052631579, + "grad_norm": 0.94140625, + "learning_rate": 0.0001644211869037027, + "loss": 1.1298, + "step": 270 + }, + { + "epoch": 1.1410526315789473, + "grad_norm": 0.341796875, + "learning_rate": 0.00016416465996504077, + "loss": 1.131, + "step": 271 + }, + { + "epoch": 1.1452631578947368, + "grad_norm": 0.90234375, + "learning_rate": 0.00016390741326330907, + "loss": 1.0148, + "step": 272 + }, + { + "epoch": 1.1494736842105264, + "grad_norm": 0.35546875, + "learning_rate": 0.00016364944968415632, + "loss": 0.788, + "step": 273 + }, + { + "epoch": 1.1536842105263159, + "grad_norm": 0.294921875, + "learning_rate": 0.00016339077212127294, + "loss": 0.9585, + "step": 274 + }, + { + "epoch": 1.1578947368421053, + "grad_norm": 0.56640625, + "learning_rate": 0.00016313138347635837, + "loss": 0.9937, + "step": 275 + }, + { + "epoch": 1.1621052631578948, + "grad_norm": 0.251953125, + "learning_rate": 0.0001628712866590885, + "loss": 1.069, + "step": 276 + }, + { + "epoch": 1.1663157894736842, + "grad_norm": 0.384765625, + "learning_rate": 0.00016261048458708318, + "loss": 1.2734, + "step": 277 + }, + { + "epoch": 1.1705263157894736, + "grad_norm": 0.66015625, + "learning_rate": 0.00016234898018587337, + "loss": 1.1955, + "step": 278 + }, + { + "epoch": 1.174736842105263, + "grad_norm": 0.3984375, + "learning_rate": 0.00016208677638886835, + "loss": 0.9949, + "step": 279 + }, + { + "epoch": 1.1789473684210527, + "grad_norm": 0.33984375, + "learning_rate": 0.00016182387613732291, + "loss": 1.1192, + "step": 280 + }, + { + "epoch": 1.1831578947368422, + "grad_norm": 0.4765625, + "learning_rate": 0.00016156028238030422, + "loss": 1.165, + "step": 281 + }, + { + "epoch": 1.1873684210526316, + "grad_norm": 0.51171875, + "learning_rate": 0.00016129599807465875, + "loss": 0.9109, + "step": 282 + }, + { + "epoch": 1.191578947368421, + "grad_norm": 0.5546875, + "learning_rate": 0.00016103102618497922, + "loss": 0.9441, + "step": 283 + }, + { + "epoch": 1.1957894736842105, + "grad_norm": 0.7109375, + "learning_rate": 0.0001607653696835713, + "loss": 1.3679, + "step": 284 + }, + { + "epoch": 1.2, + "grad_norm": 0.3671875, + "learning_rate": 0.0001604990315504202, + "loss": 0.8203, + "step": 285 + }, + { + "epoch": 1.2042105263157894, + "grad_norm": 0.51171875, + "learning_rate": 0.00016023201477315731, + "loss": 1.052, + "step": 286 + }, + { + "epoch": 1.208421052631579, + "grad_norm": 0.72265625, + "learning_rate": 0.00015996432234702666, + "loss": 1.281, + "step": 287 + }, + { + "epoch": 1.2126315789473685, + "grad_norm": 0.44921875, + "learning_rate": 0.0001596959572748514, + "loss": 1.1851, + "step": 288 + }, + { + "epoch": 1.216842105263158, + "grad_norm": 1.3515625, + "learning_rate": 0.0001594269225669999, + "loss": 1.1378, + "step": 289 + }, + { + "epoch": 1.2210526315789474, + "grad_norm": 0.337890625, + "learning_rate": 0.00015915722124135227, + "loss": 0.7495, + "step": 290 + }, + { + "epoch": 1.2252631578947368, + "grad_norm": 0.45703125, + "learning_rate": 0.00015888685632326627, + "loss": 1.0089, + "step": 291 + }, + { + "epoch": 1.2294736842105263, + "grad_norm": 0.470703125, + "learning_rate": 0.00015861583084554349, + "loss": 0.7145, + "step": 292 + }, + { + "epoch": 1.2336842105263157, + "grad_norm": 0.4375, + "learning_rate": 0.00015834414784839532, + "loss": 1.1728, + "step": 293 + }, + { + "epoch": 1.2378947368421054, + "grad_norm": 1.5703125, + "learning_rate": 0.0001580718103794089, + "loss": 1.351, + "step": 294 + }, + { + "epoch": 1.2421052631578948, + "grad_norm": 0.6171875, + "learning_rate": 0.0001577988214935127, + "loss": 1.0116, + "step": 295 + }, + { + "epoch": 1.2463157894736843, + "grad_norm": 0.392578125, + "learning_rate": 0.00015752518425294257, + "loss": 1.1591, + "step": 296 + }, + { + "epoch": 1.2505263157894737, + "grad_norm": 0.44140625, + "learning_rate": 0.0001572509017272072, + "loss": 1.0848, + "step": 297 + }, + { + "epoch": 1.2547368421052632, + "grad_norm": 0.43359375, + "learning_rate": 0.00015697597699305366, + "loss": 1.0024, + "step": 298 + }, + { + "epoch": 1.2589473684210526, + "grad_norm": 0.470703125, + "learning_rate": 0.00015670041313443304, + "loss": 1.072, + "step": 299 + }, + { + "epoch": 1.263157894736842, + "grad_norm": 0.34375, + "learning_rate": 0.00015642421324246568, + "loss": 1.1825, + "step": 300 + }, + { + "epoch": 1.263157894736842, + "eval_loss": 1.207429051399231, + "eval_runtime": 25.9189, + "eval_samples_per_second": 3.858, + "eval_steps_per_second": 1.929, + "step": 300 + }, + { + "epoch": 1.2673684210526317, + "grad_norm": 1.0234375, + "learning_rate": 0.00015614738041540668, + "loss": 0.8984, + "step": 301 + }, + { + "epoch": 1.271578947368421, + "grad_norm": 0.29296875, + "learning_rate": 0.00015586991775861102, + "loss": 0.9569, + "step": 302 + }, + { + "epoch": 1.2757894736842106, + "grad_norm": 0.59375, + "learning_rate": 0.00015559182838449873, + "loss": 1.177, + "step": 303 + }, + { + "epoch": 1.28, + "grad_norm": 0.416015625, + "learning_rate": 0.00015531311541251995, + "loss": 0.9957, + "step": 304 + }, + { + "epoch": 1.2842105263157895, + "grad_norm": 0.47265625, + "learning_rate": 0.00015503378196912013, + "loss": 0.8083, + "step": 305 + }, + { + "epoch": 1.288421052631579, + "grad_norm": 0.7734375, + "learning_rate": 0.00015475383118770472, + "loss": 1.0463, + "step": 306 + }, + { + "epoch": 1.2926315789473684, + "grad_norm": 0.30859375, + "learning_rate": 0.00015447326620860415, + "loss": 1.0709, + "step": 307 + }, + { + "epoch": 1.296842105263158, + "grad_norm": 0.35546875, + "learning_rate": 0.00015419209017903852, + "loss": 1.2547, + "step": 308 + }, + { + "epoch": 1.3010526315789472, + "grad_norm": 0.3671875, + "learning_rate": 0.00015391030625308249, + "loss": 0.8027, + "step": 309 + }, + { + "epoch": 1.305263157894737, + "grad_norm": 0.28515625, + "learning_rate": 0.0001536279175916296, + "loss": 1.0735, + "step": 310 + }, + { + "epoch": 1.3094736842105263, + "grad_norm": 0.625, + "learning_rate": 0.00015334492736235705, + "loss": 1.0046, + "step": 311 + }, + { + "epoch": 1.3136842105263158, + "grad_norm": 0.83203125, + "learning_rate": 0.0001530613387396901, + "loss": 1.1327, + "step": 312 + }, + { + "epoch": 1.3178947368421052, + "grad_norm": 0.515625, + "learning_rate": 0.00015277715490476641, + "loss": 0.9917, + "step": 313 + }, + { + "epoch": 1.3221052631578947, + "grad_norm": 0.359375, + "learning_rate": 0.0001524923790454004, + "loss": 1.2532, + "step": 314 + }, + { + "epoch": 1.3263157894736843, + "grad_norm": 0.306640625, + "learning_rate": 0.0001522070143560476, + "loss": 0.9906, + "step": 315 + }, + { + "epoch": 1.3305263157894736, + "grad_norm": 0.33984375, + "learning_rate": 0.00015192106403776848, + "loss": 1.0958, + "step": 316 + }, + { + "epoch": 1.3347368421052632, + "grad_norm": 0.33203125, + "learning_rate": 0.000151634531298193, + "loss": 1.1674, + "step": 317 + }, + { + "epoch": 1.3389473684210527, + "grad_norm": 0.45703125, + "learning_rate": 0.0001513474193514842, + "loss": 0.9189, + "step": 318 + }, + { + "epoch": 1.343157894736842, + "grad_norm": 0.55078125, + "learning_rate": 0.0001510597314183025, + "loss": 1.2956, + "step": 319 + }, + { + "epoch": 1.3473684210526315, + "grad_norm": 0.41015625, + "learning_rate": 0.00015077147072576933, + "loss": 1.3053, + "step": 320 + }, + { + "epoch": 1.351578947368421, + "grad_norm": 0.421875, + "learning_rate": 0.00015048264050743096, + "loss": 1.26, + "step": 321 + }, + { + "epoch": 1.3557894736842107, + "grad_norm": 0.318359375, + "learning_rate": 0.00015019324400322243, + "loss": 1.1314, + "step": 322 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 0.33203125, + "learning_rate": 0.00014990328445943092, + "loss": 1.1536, + "step": 323 + }, + { + "epoch": 1.3642105263157895, + "grad_norm": 0.365234375, + "learning_rate": 0.00014961276512865954, + "loss": 0.8678, + "step": 324 + }, + { + "epoch": 1.368421052631579, + "grad_norm": 0.56640625, + "learning_rate": 0.00014932168926979074, + "loss": 1.0508, + "step": 325 + }, + { + "epoch": 1.3726315789473684, + "grad_norm": 0.333984375, + "learning_rate": 0.00014903006014794983, + "loss": 0.9275, + "step": 326 + }, + { + "epoch": 1.3768421052631579, + "grad_norm": 0.61328125, + "learning_rate": 0.00014873788103446825, + "loss": 1.0225, + "step": 327 + }, + { + "epoch": 1.3810526315789473, + "grad_norm": 0.357421875, + "learning_rate": 0.00014844515520684703, + "loss": 0.9955, + "step": 328 + }, + { + "epoch": 1.385263157894737, + "grad_norm": 0.6328125, + "learning_rate": 0.00014815188594871983, + "loss": 1.2857, + "step": 329 + }, + { + "epoch": 1.3894736842105262, + "grad_norm": 0.5078125, + "learning_rate": 0.00014785807654981627, + "loss": 1.0598, + "step": 330 + }, + { + "epoch": 1.3936842105263159, + "grad_norm": 0.96484375, + "learning_rate": 0.00014756373030592497, + "loss": 1.1031, + "step": 331 + }, + { + "epoch": 1.3978947368421053, + "grad_norm": 0.380859375, + "learning_rate": 0.00014726885051885653, + "loss": 1.3287, + "step": 332 + }, + { + "epoch": 1.4021052631578947, + "grad_norm": 0.5859375, + "learning_rate": 0.00014697344049640658, + "loss": 1.1952, + "step": 333 + }, + { + "epoch": 1.4063157894736842, + "grad_norm": 0.453125, + "learning_rate": 0.0001466775035523186, + "loss": 1.083, + "step": 334 + }, + { + "epoch": 1.4105263157894736, + "grad_norm": 0.37890625, + "learning_rate": 0.00014638104300624685, + "loss": 0.8393, + "step": 335 + }, + { + "epoch": 1.4147368421052633, + "grad_norm": 0.9921875, + "learning_rate": 0.00014608406218371894, + "loss": 1.3947, + "step": 336 + }, + { + "epoch": 1.4189473684210525, + "grad_norm": 0.357421875, + "learning_rate": 0.0001457865644160987, + "loss": 1.028, + "step": 337 + }, + { + "epoch": 1.4231578947368422, + "grad_norm": 0.671875, + "learning_rate": 0.00014548855304054886, + "loss": 1.1156, + "step": 338 + }, + { + "epoch": 1.4273684210526316, + "grad_norm": 0.6953125, + "learning_rate": 0.00014519003139999337, + "loss": 1.1537, + "step": 339 + }, + { + "epoch": 1.431578947368421, + "grad_norm": 0.408203125, + "learning_rate": 0.00014489100284308017, + "loss": 1.207, + "step": 340 + }, + { + "epoch": 1.4357894736842105, + "grad_norm": 0.56640625, + "learning_rate": 0.00014459147072414335, + "loss": 1.125, + "step": 341 + }, + { + "epoch": 1.44, + "grad_norm": 0.423828125, + "learning_rate": 0.00014429143840316585, + "loss": 1.0774, + "step": 342 + }, + { + "epoch": 1.4442105263157896, + "grad_norm": 1.0859375, + "learning_rate": 0.0001439909092457415, + "loss": 1.3349, + "step": 343 + }, + { + "epoch": 1.4484210526315788, + "grad_norm": 0.55078125, + "learning_rate": 0.00014368988662303732, + "loss": 0.6138, + "step": 344 + }, + { + "epoch": 1.4526315789473685, + "grad_norm": 0.330078125, + "learning_rate": 0.00014338837391175582, + "loss": 1.1961, + "step": 345 + }, + { + "epoch": 1.456842105263158, + "grad_norm": 0.54296875, + "learning_rate": 0.00014308637449409706, + "loss": 1.067, + "step": 346 + }, + { + "epoch": 1.4610526315789474, + "grad_norm": 0.294921875, + "learning_rate": 0.0001427838917577206, + "loss": 1.0033, + "step": 347 + }, + { + "epoch": 1.4652631578947368, + "grad_norm": 0.494140625, + "learning_rate": 0.00014248092909570774, + "loss": 1.011, + "step": 348 + }, + { + "epoch": 1.4694736842105263, + "grad_norm": 0.75, + "learning_rate": 0.00014217748990652318, + "loss": 1.3504, + "step": 349 + }, + { + "epoch": 1.4736842105263157, + "grad_norm": 0.41796875, + "learning_rate": 0.00014187357759397714, + "loss": 1.0421, + "step": 350 + }, + { + "epoch": 1.4778947368421052, + "grad_norm": 0.73828125, + "learning_rate": 0.00014156919556718698, + "loss": 1.1829, + "step": 351 + }, + { + "epoch": 1.4821052631578948, + "grad_norm": 0.31640625, + "learning_rate": 0.00014126434724053913, + "loss": 1.1985, + "step": 352 + }, + { + "epoch": 1.4863157894736843, + "grad_norm": 0.41015625, + "learning_rate": 0.00014095903603365066, + "loss": 1.1609, + "step": 353 + }, + { + "epoch": 1.4905263157894737, + "grad_norm": 0.345703125, + "learning_rate": 0.00014065326537133094, + "loss": 1.2552, + "step": 354 + }, + { + "epoch": 1.4947368421052631, + "grad_norm": 1.2265625, + "learning_rate": 0.0001403470386835434, + "loss": 0.9655, + "step": 355 + }, + { + "epoch": 1.4989473684210526, + "grad_norm": 0.46875, + "learning_rate": 0.0001400403594053667, + "loss": 0.9709, + "step": 356 + }, + { + "epoch": 1.5031578947368422, + "grad_norm": 0.703125, + "learning_rate": 0.00013973323097695657, + "loss": 1.048, + "step": 357 + }, + { + "epoch": 1.5073684210526315, + "grad_norm": 0.494140625, + "learning_rate": 0.00013942565684350698, + "loss": 1.0853, + "step": 358 + }, + { + "epoch": 1.5115789473684211, + "grad_norm": 0.9140625, + "learning_rate": 0.00013911764045521159, + "loss": 1.1191, + "step": 359 + }, + { + "epoch": 1.5157894736842106, + "grad_norm": 0.478515625, + "learning_rate": 0.00013880918526722497, + "loss": 0.9782, + "step": 360 + }, + { + "epoch": 1.5157894736842106, + "eval_loss": 1.2060081958770752, + "eval_runtime": 25.9425, + "eval_samples_per_second": 3.855, + "eval_steps_per_second": 1.927, + "step": 360 + }, + { + "epoch": 1.52, + "grad_norm": 0.361328125, + "learning_rate": 0.00013850029473962402, + "loss": 0.9822, + "step": 361 + }, + { + "epoch": 1.5242105263157895, + "grad_norm": 0.44921875, + "learning_rate": 0.00013819097233736888, + "loss": 0.957, + "step": 362 + }, + { + "epoch": 1.528421052631579, + "grad_norm": 0.478515625, + "learning_rate": 0.0001378812215302644, + "loss": 0.9774, + "step": 363 + }, + { + "epoch": 1.5326315789473686, + "grad_norm": 0.35546875, + "learning_rate": 0.00013757104579292082, + "loss": 1.2854, + "step": 364 + }, + { + "epoch": 1.5368421052631578, + "grad_norm": 0.671875, + "learning_rate": 0.00013726044860471522, + "loss": 0.9209, + "step": 365 + }, + { + "epoch": 1.5410526315789475, + "grad_norm": 0.400390625, + "learning_rate": 0.00013694943344975212, + "loss": 0.9489, + "step": 366 + }, + { + "epoch": 1.545263157894737, + "grad_norm": 0.625, + "learning_rate": 0.00013663800381682464, + "loss": 1.2956, + "step": 367 + }, + { + "epoch": 1.5494736842105263, + "grad_norm": 0.3203125, + "learning_rate": 0.00013632616319937522, + "loss": 1.0642, + "step": 368 + }, + { + "epoch": 1.5536842105263158, + "grad_norm": 0.296875, + "learning_rate": 0.0001360139150954566, + "loss": 1.034, + "step": 369 + }, + { + "epoch": 1.5578947368421052, + "grad_norm": 0.5234375, + "learning_rate": 0.00013570126300769232, + "loss": 0.8314, + "step": 370 + }, + { + "epoch": 1.5621052631578949, + "grad_norm": 0.279296875, + "learning_rate": 0.00013538821044323772, + "loss": 1.1934, + "step": 371 + }, + { + "epoch": 1.566315789473684, + "grad_norm": 0.353515625, + "learning_rate": 0.0001350747609137404, + "loss": 1.2661, + "step": 372 + }, + { + "epoch": 1.5705263157894738, + "grad_norm": 0.27734375, + "learning_rate": 0.00013476091793530092, + "loss": 0.9746, + "step": 373 + }, + { + "epoch": 1.5747368421052632, + "grad_norm": 0.49609375, + "learning_rate": 0.0001344466850284333, + "loss": 1.0562, + "step": 374 + }, + { + "epoch": 1.5789473684210527, + "grad_norm": 0.27734375, + "learning_rate": 0.00013413206571802558, + "loss": 0.8612, + "step": 375 + }, + { + "epoch": 1.583157894736842, + "grad_norm": 0.51953125, + "learning_rate": 0.00013381706353330014, + "loss": 0.7912, + "step": 376 + }, + { + "epoch": 1.5873684210526315, + "grad_norm": 0.45703125, + "learning_rate": 0.00013350168200777448, + "loss": 1.0244, + "step": 377 + }, + { + "epoch": 1.5915789473684212, + "grad_norm": 0.5703125, + "learning_rate": 0.0001331859246792211, + "loss": 1.1267, + "step": 378 + }, + { + "epoch": 1.5957894736842104, + "grad_norm": 0.48828125, + "learning_rate": 0.00013286979508962815, + "loss": 1.0055, + "step": 379 + }, + { + "epoch": 1.6, + "grad_norm": 0.310546875, + "learning_rate": 0.0001325532967851596, + "loss": 1.2348, + "step": 380 + }, + { + "epoch": 1.6042105263157893, + "grad_norm": 0.369140625, + "learning_rate": 0.00013223643331611537, + "loss": 0.991, + "step": 381 + }, + { + "epoch": 1.608421052631579, + "grad_norm": 0.51171875, + "learning_rate": 0.00013191920823689177, + "loss": 1.0261, + "step": 382 + }, + { + "epoch": 1.6126315789473684, + "grad_norm": 0.90234375, + "learning_rate": 0.00013160162510594123, + "loss": 1.0752, + "step": 383 + }, + { + "epoch": 1.6168421052631579, + "grad_norm": 1.1796875, + "learning_rate": 0.00013128368748573273, + "loss": 1.1808, + "step": 384 + }, + { + "epoch": 1.6210526315789475, + "grad_norm": 0.40625, + "learning_rate": 0.0001309653989427117, + "loss": 0.8969, + "step": 385 + }, + { + "epoch": 1.6252631578947367, + "grad_norm": 0.53515625, + "learning_rate": 0.00013064676304726, + "loss": 1.1139, + "step": 386 + }, + { + "epoch": 1.6294736842105264, + "grad_norm": 0.478515625, + "learning_rate": 0.00013032778337365592, + "loss": 1.2781, + "step": 387 + }, + { + "epoch": 1.6336842105263156, + "grad_norm": 0.66796875, + "learning_rate": 0.0001300084635000341, + "loss": 1.1434, + "step": 388 + }, + { + "epoch": 1.6378947368421053, + "grad_norm": 0.33984375, + "learning_rate": 0.00012968880700834517, + "loss": 0.7428, + "step": 389 + }, + { + "epoch": 1.6421052631578947, + "grad_norm": 0.357421875, + "learning_rate": 0.000129368817484316, + "loss": 1.256, + "step": 390 + }, + { + "epoch": 1.6463157894736842, + "grad_norm": 0.384765625, + "learning_rate": 0.00012904849851740906, + "loss": 1.1516, + "step": 391 + }, + { + "epoch": 1.6505263157894738, + "grad_norm": 0.62890625, + "learning_rate": 0.0001287278537007824, + "loss": 1.1683, + "step": 392 + }, + { + "epoch": 1.654736842105263, + "grad_norm": 0.4140625, + "learning_rate": 0.0001284068866312492, + "loss": 0.9682, + "step": 393 + }, + { + "epoch": 1.6589473684210527, + "grad_norm": 0.36328125, + "learning_rate": 0.00012808560090923758, + "loss": 1.2109, + "step": 394 + }, + { + "epoch": 1.663157894736842, + "grad_norm": 0.3125, + "learning_rate": 0.00012776400013875006, + "loss": 1.1139, + "step": 395 + }, + { + "epoch": 1.6673684210526316, + "grad_norm": 0.453125, + "learning_rate": 0.00012744208792732324, + "loss": 1.0977, + "step": 396 + }, + { + "epoch": 1.671578947368421, + "grad_norm": 0.44140625, + "learning_rate": 0.00012711986788598725, + "loss": 1.1872, + "step": 397 + }, + { + "epoch": 1.6757894736842105, + "grad_norm": 0.59765625, + "learning_rate": 0.00012679734362922528, + "loss": 1.0773, + "step": 398 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 0.265625, + "learning_rate": 0.00012647451877493307, + "loss": 0.6203, + "step": 399 + }, + { + "epoch": 1.6842105263157894, + "grad_norm": 0.5078125, + "learning_rate": 0.00012615139694437835, + "loss": 1.1557, + "step": 400 + }, + { + "epoch": 1.688421052631579, + "grad_norm": 0.578125, + "learning_rate": 0.00012582798176215997, + "loss": 1.1022, + "step": 401 + }, + { + "epoch": 1.6926315789473683, + "grad_norm": 0.4609375, + "learning_rate": 0.00012550427685616765, + "loss": 1.193, + "step": 402 + }, + { + "epoch": 1.696842105263158, + "grad_norm": 0.455078125, + "learning_rate": 0.00012518028585754094, + "loss": 0.8758, + "step": 403 + }, + { + "epoch": 1.7010526315789474, + "grad_norm": 0.51171875, + "learning_rate": 0.00012485601240062869, + "loss": 0.9792, + "step": 404 + }, + { + "epoch": 1.7052631578947368, + "grad_norm": 0.3046875, + "learning_rate": 0.00012453146012294812, + "loss": 1.1347, + "step": 405 + }, + { + "epoch": 1.7094736842105265, + "grad_norm": 0.439453125, + "learning_rate": 0.00012420663266514417, + "loss": 1.1692, + "step": 406 + }, + { + "epoch": 1.7136842105263157, + "grad_norm": 0.33203125, + "learning_rate": 0.00012388153367094858, + "loss": 0.9904, + "step": 407 + }, + { + "epoch": 1.7178947368421054, + "grad_norm": 0.62109375, + "learning_rate": 0.0001235561667871391, + "loss": 0.8854, + "step": 408 + }, + { + "epoch": 1.7221052631578946, + "grad_norm": 0.52734375, + "learning_rate": 0.00012323053566349834, + "loss": 1.3143, + "step": 409 + }, + { + "epoch": 1.7263157894736842, + "grad_norm": 0.390625, + "learning_rate": 0.0001229046439527732, + "loss": 1.0957, + "step": 410 + }, + { + "epoch": 1.7305263157894737, + "grad_norm": 0.494140625, + "learning_rate": 0.00012257849531063352, + "loss": 1.0458, + "step": 411 + }, + { + "epoch": 1.7347368421052631, + "grad_norm": 0.34375, + "learning_rate": 0.00012225209339563145, + "loss": 1.3262, + "step": 412 + }, + { + "epoch": 1.7389473684210528, + "grad_norm": 0.32421875, + "learning_rate": 0.00012192544186916006, + "loss": 1.1506, + "step": 413 + }, + { + "epoch": 1.743157894736842, + "grad_norm": 0.671875, + "learning_rate": 0.00012159854439541245, + "loss": 1.035, + "step": 414 + }, + { + "epoch": 1.7473684210526317, + "grad_norm": 0.4453125, + "learning_rate": 0.00012127140464134066, + "loss": 1.3485, + "step": 415 + }, + { + "epoch": 1.751578947368421, + "grad_norm": 0.439453125, + "learning_rate": 0.00012094402627661447, + "loss": 1.1842, + "step": 416 + }, + { + "epoch": 1.7557894736842106, + "grad_norm": 0.51171875, + "learning_rate": 0.00012061641297358022, + "loss": 1.4594, + "step": 417 + }, + { + "epoch": 1.76, + "grad_norm": 0.34375, + "learning_rate": 0.00012028856840721974, + "loss": 1.1014, + "step": 418 + }, + { + "epoch": 1.7642105263157895, + "grad_norm": 0.640625, + "learning_rate": 0.00011996049625510901, + "loss": 0.7333, + "step": 419 + }, + { + "epoch": 1.768421052631579, + "grad_norm": 0.5859375, + "learning_rate": 0.00011963220019737691, + "loss": 1.2063, + "step": 420 + }, + { + "epoch": 1.768421052631579, + "eval_loss": 1.199410080909729, + "eval_runtime": 25.8809, + "eval_samples_per_second": 3.864, + "eval_steps_per_second": 1.932, + "step": 420 + }, + { + "epoch": 1.7726315789473683, + "grad_norm": 0.73046875, + "learning_rate": 0.00011930368391666398, + "loss": 1.2112, + "step": 421 + }, + { + "epoch": 1.776842105263158, + "grad_norm": 0.67578125, + "learning_rate": 0.00011897495109808107, + "loss": 1.0342, + "step": 422 + }, + { + "epoch": 1.7810526315789472, + "grad_norm": 0.376953125, + "learning_rate": 0.00011864600542916813, + "loss": 1.2239, + "step": 423 + }, + { + "epoch": 1.7852631578947369, + "grad_norm": 0.361328125, + "learning_rate": 0.00011831685059985262, + "loss": 0.9817, + "step": 424 + }, + { + "epoch": 1.7894736842105263, + "grad_norm": 0.30859375, + "learning_rate": 0.00011798749030240834, + "loss": 1.3617, + "step": 425 + }, + { + "epoch": 1.7936842105263158, + "grad_norm": 0.7265625, + "learning_rate": 0.00011765792823141384, + "loss": 1.2746, + "step": 426 + }, + { + "epoch": 1.7978947368421052, + "grad_norm": 0.484375, + "learning_rate": 0.00011732816808371115, + "loss": 1.471, + "step": 427 + }, + { + "epoch": 1.8021052631578947, + "grad_norm": 0.349609375, + "learning_rate": 0.00011699821355836409, + "loss": 1.1071, + "step": 428 + }, + { + "epoch": 1.8063157894736843, + "grad_norm": 0.26953125, + "learning_rate": 0.00011666806835661699, + "loss": 1.0702, + "step": 429 + }, + { + "epoch": 1.8105263157894735, + "grad_norm": 0.31640625, + "learning_rate": 0.00011633773618185302, + "loss": 1.2166, + "step": 430 + }, + { + "epoch": 1.8147368421052632, + "grad_norm": 0.36328125, + "learning_rate": 0.00011600722073955282, + "loss": 1.1926, + "step": 431 + }, + { + "epoch": 1.8189473684210526, + "grad_norm": 0.35546875, + "learning_rate": 0.00011567652573725262, + "loss": 1.0472, + "step": 432 + }, + { + "epoch": 1.823157894736842, + "grad_norm": 0.765625, + "learning_rate": 0.00011534565488450311, + "loss": 0.9258, + "step": 433 + }, + { + "epoch": 1.8273684210526315, + "grad_norm": 0.765625, + "learning_rate": 0.00011501461189282733, + "loss": 1.6986, + "step": 434 + }, + { + "epoch": 1.831578947368421, + "grad_norm": 0.392578125, + "learning_rate": 0.00011468340047567947, + "loss": 1.1726, + "step": 435 + }, + { + "epoch": 1.8357894736842106, + "grad_norm": 0.412109375, + "learning_rate": 0.00011435202434840287, + "loss": 1.2373, + "step": 436 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 0.6015625, + "learning_rate": 0.0001140204872281886, + "loss": 0.7262, + "step": 437 + }, + { + "epoch": 1.8442105263157895, + "grad_norm": 0.490234375, + "learning_rate": 0.0001136887928340336, + "loss": 1.0527, + "step": 438 + }, + { + "epoch": 1.848421052631579, + "grad_norm": 0.453125, + "learning_rate": 0.00011335694488669904, + "loss": 0.89, + "step": 439 + }, + { + "epoch": 1.8526315789473684, + "grad_norm": 0.55859375, + "learning_rate": 0.00011302494710866857, + "loss": 1.109, + "step": 440 + }, + { + "epoch": 1.8568421052631578, + "grad_norm": 0.5546875, + "learning_rate": 0.00011269280322410658, + "loss": 1.0818, + "step": 441 + }, + { + "epoch": 1.8610526315789473, + "grad_norm": 0.3671875, + "learning_rate": 0.00011236051695881633, + "loss": 1.2366, + "step": 442 + }, + { + "epoch": 1.865263157894737, + "grad_norm": 0.427734375, + "learning_rate": 0.0001120280920401983, + "loss": 1.1401, + "step": 443 + }, + { + "epoch": 1.8694736842105262, + "grad_norm": 0.5, + "learning_rate": 0.00011169553219720828, + "loss": 1.0061, + "step": 444 + }, + { + "epoch": 1.8736842105263158, + "grad_norm": 0.43359375, + "learning_rate": 0.0001113628411603155, + "loss": 1.2742, + "step": 445 + }, + { + "epoch": 1.8778947368421053, + "grad_norm": 0.375, + "learning_rate": 0.00011103002266146096, + "loss": 1.296, + "step": 446 + }, + { + "epoch": 1.8821052631578947, + "grad_norm": 0.7578125, + "learning_rate": 0.0001106970804340154, + "loss": 0.9062, + "step": 447 + }, + { + "epoch": 1.8863157894736842, + "grad_norm": 0.44140625, + "learning_rate": 0.0001103640182127375, + "loss": 0.85, + "step": 448 + }, + { + "epoch": 1.8905263157894736, + "grad_norm": 0.6171875, + "learning_rate": 0.00011003083973373187, + "loss": 1.0045, + "step": 449 + }, + { + "epoch": 1.8947368421052633, + "grad_norm": 0.50390625, + "learning_rate": 0.00010969754873440743, + "loss": 0.8616, + "step": 450 + }, + { + "epoch": 1.8989473684210525, + "grad_norm": 0.5546875, + "learning_rate": 0.0001093641489534351, + "loss": 0.9867, + "step": 451 + }, + { + "epoch": 1.9031578947368422, + "grad_norm": 0.462890625, + "learning_rate": 0.00010903064413070612, + "loss": 1.1836, + "step": 452 + }, + { + "epoch": 1.9073684210526316, + "grad_norm": 0.515625, + "learning_rate": 0.00010869703800729, + "loss": 0.9996, + "step": 453 + }, + { + "epoch": 1.911578947368421, + "grad_norm": 0.5078125, + "learning_rate": 0.00010836333432539272, + "loss": 1.4083, + "step": 454 + }, + { + "epoch": 1.9157894736842105, + "grad_norm": 0.474609375, + "learning_rate": 0.0001080295368283144, + "loss": 0.9889, + "step": 455 + }, + { + "epoch": 1.92, + "grad_norm": 0.30078125, + "learning_rate": 0.00010769564926040769, + "loss": 0.8322, + "step": 456 + }, + { + "epoch": 1.9242105263157896, + "grad_norm": 1.171875, + "learning_rate": 0.00010736167536703556, + "loss": 0.8793, + "step": 457 + }, + { + "epoch": 1.9284210526315788, + "grad_norm": 0.59765625, + "learning_rate": 0.0001070276188945293, + "loss": 1.1383, + "step": 458 + }, + { + "epoch": 1.9326315789473685, + "grad_norm": 0.625, + "learning_rate": 0.00010669348359014657, + "loss": 1.2608, + "step": 459 + }, + { + "epoch": 1.936842105263158, + "grad_norm": 0.49609375, + "learning_rate": 0.00010635927320202928, + "loss": 1.1444, + "step": 460 + }, + { + "epoch": 1.9410526315789474, + "grad_norm": 0.6796875, + "learning_rate": 0.00010602499147916159, + "loss": 1.1779, + "step": 461 + }, + { + "epoch": 1.9452631578947368, + "grad_norm": 0.390625, + "learning_rate": 0.00010569064217132791, + "loss": 1.2066, + "step": 462 + }, + { + "epoch": 1.9494736842105262, + "grad_norm": 0.32421875, + "learning_rate": 0.00010535622902907072, + "loss": 1.3065, + "step": 463 + }, + { + "epoch": 1.953684210526316, + "grad_norm": 0.45703125, + "learning_rate": 0.00010502175580364857, + "loss": 0.9167, + "step": 464 + }, + { + "epoch": 1.9578947368421051, + "grad_norm": 0.8203125, + "learning_rate": 0.00010468722624699401, + "loss": 1.059, + "step": 465 + }, + { + "epoch": 1.9621052631578948, + "grad_norm": 0.93359375, + "learning_rate": 0.00010435264411167148, + "loss": 0.8437, + "step": 466 + }, + { + "epoch": 1.9663157894736842, + "grad_norm": 0.4453125, + "learning_rate": 0.00010401801315083519, + "loss": 1.1105, + "step": 467 + }, + { + "epoch": 1.9705263157894737, + "grad_norm": 0.81640625, + "learning_rate": 0.0001036833371181871, + "loss": 1.5699, + "step": 468 + }, + { + "epoch": 1.9747368421052631, + "grad_norm": 0.359375, + "learning_rate": 0.00010334861976793465, + "loss": 1.0942, + "step": 469 + }, + { + "epoch": 1.9789473684210526, + "grad_norm": 0.4765625, + "learning_rate": 0.00010301386485474889, + "loss": 0.923, + "step": 470 + }, + { + "epoch": 1.9831578947368422, + "grad_norm": 0.498046875, + "learning_rate": 0.00010267907613372217, + "loss": 1.123, + "step": 471 + }, + { + "epoch": 1.9873684210526315, + "grad_norm": 0.45703125, + "learning_rate": 0.00010234425736032607, + "loss": 1.1807, + "step": 472 + }, + { + "epoch": 1.9915789473684211, + "grad_norm": 0.53515625, + "learning_rate": 0.00010200941229036926, + "loss": 1.2465, + "step": 473 + }, + { + "epoch": 1.9957894736842106, + "grad_norm": 0.3984375, + "learning_rate": 0.00010167454467995549, + "loss": 0.9394, + "step": 474 + }, + { + "epoch": 2.0, + "grad_norm": 1.0625, + "learning_rate": 0.00010133965828544128, + "loss": 1.0323, + "step": 475 + }, + { + "epoch": 2.0042105263157897, + "grad_norm": 0.318359375, + "learning_rate": 0.00010100475686339379, + "loss": 1.1268, + "step": 476 + }, + { + "epoch": 2.008421052631579, + "grad_norm": 0.416015625, + "learning_rate": 0.00010066984417054887, + "loss": 0.7652, + "step": 477 + }, + { + "epoch": 2.0126315789473685, + "grad_norm": 0.41015625, + "learning_rate": 0.00010033492396376878, + "loss": 1.0744, + "step": 478 + }, + { + "epoch": 2.0168421052631578, + "grad_norm": 0.296875, + "learning_rate": 0.0001, + "loss": 1.0926, + "step": 479 + }, + { + "epoch": 2.0210526315789474, + "grad_norm": 0.3671875, + "learning_rate": 9.966507603623125e-05, + "loss": 0.9614, + "step": 480 + }, + { + "epoch": 2.0210526315789474, + "eval_loss": 1.1929166316986084, + "eval_runtime": 25.8683, + "eval_samples_per_second": 3.866, + "eval_steps_per_second": 1.933, + "step": 480 + }, + { + "epoch": 2.0252631578947367, + "grad_norm": 0.48046875, + "learning_rate": 9.933015582945112e-05, + "loss": 1.0848, + "step": 481 + }, + { + "epoch": 2.0294736842105263, + "grad_norm": 0.361328125, + "learning_rate": 9.899524313660624e-05, + "loss": 1.0067, + "step": 482 + }, + { + "epoch": 2.033684210526316, + "grad_norm": 0.4375, + "learning_rate": 9.866034171455876e-05, + "loss": 0.7833, + "step": 483 + }, + { + "epoch": 2.037894736842105, + "grad_norm": 0.33984375, + "learning_rate": 9.832545532004454e-05, + "loss": 0.9732, + "step": 484 + }, + { + "epoch": 2.042105263157895, + "grad_norm": 0.341796875, + "learning_rate": 9.799058770963073e-05, + "loss": 0.6746, + "step": 485 + }, + { + "epoch": 2.046315789473684, + "grad_norm": 0.439453125, + "learning_rate": 9.765574263967396e-05, + "loss": 0.7767, + "step": 486 + }, + { + "epoch": 2.0505263157894738, + "grad_norm": 0.54296875, + "learning_rate": 9.732092386627787e-05, + "loss": 0.7498, + "step": 487 + }, + { + "epoch": 2.054736842105263, + "grad_norm": 0.55859375, + "learning_rate": 9.698613514525116e-05, + "loss": 0.6573, + "step": 488 + }, + { + "epoch": 2.0589473684210526, + "grad_norm": 0.40234375, + "learning_rate": 9.665138023206536e-05, + "loss": 0.6588, + "step": 489 + }, + { + "epoch": 2.0631578947368423, + "grad_norm": 0.37109375, + "learning_rate": 9.631666288181293e-05, + "loss": 0.9873, + "step": 490 + }, + { + "epoch": 2.0673684210526315, + "grad_norm": 0.44921875, + "learning_rate": 9.598198684916483e-05, + "loss": 0.7607, + "step": 491 + }, + { + "epoch": 2.071578947368421, + "grad_norm": 0.41015625, + "learning_rate": 9.564735588832856e-05, + "loss": 1.1862, + "step": 492 + }, + { + "epoch": 2.0757894736842104, + "grad_norm": 0.6328125, + "learning_rate": 9.531277375300599e-05, + "loss": 0.9608, + "step": 493 + }, + { + "epoch": 2.08, + "grad_norm": 0.82421875, + "learning_rate": 9.497824419635144e-05, + "loss": 0.6041, + "step": 494 + }, + { + "epoch": 2.0842105263157893, + "grad_norm": 0.60546875, + "learning_rate": 9.464377097092932e-05, + "loss": 1.4274, + "step": 495 + }, + { + "epoch": 2.088421052631579, + "grad_norm": 0.8671875, + "learning_rate": 9.430935782867212e-05, + "loss": 0.6098, + "step": 496 + }, + { + "epoch": 2.0926315789473686, + "grad_norm": 0.58984375, + "learning_rate": 9.397500852083841e-05, + "loss": 1.0759, + "step": 497 + }, + { + "epoch": 2.096842105263158, + "grad_norm": 0.5546875, + "learning_rate": 9.364072679797073e-05, + "loss": 0.8507, + "step": 498 + }, + { + "epoch": 2.1010526315789475, + "grad_norm": 1.09375, + "learning_rate": 9.330651640985346e-05, + "loss": 1.2054, + "step": 499 + }, + { + "epoch": 2.1052631578947367, + "grad_norm": 0.490234375, + "learning_rate": 9.297238110547074e-05, + "loss": 1.0956, + "step": 500 + }, + { + "epoch": 2.1094736842105264, + "grad_norm": 0.482421875, + "learning_rate": 9.263832463296445e-05, + "loss": 1.0093, + "step": 501 + }, + { + "epoch": 2.1136842105263156, + "grad_norm": 0.73046875, + "learning_rate": 9.230435073959232e-05, + "loss": 1.0679, + "step": 502 + }, + { + "epoch": 2.1178947368421053, + "grad_norm": 0.54296875, + "learning_rate": 9.197046317168563e-05, + "loss": 0.8422, + "step": 503 + }, + { + "epoch": 2.122105263157895, + "grad_norm": 0.66015625, + "learning_rate": 9.163666567460733e-05, + "loss": 0.6545, + "step": 504 + }, + { + "epoch": 2.126315789473684, + "grad_norm": 0.578125, + "learning_rate": 9.130296199270999e-05, + "loss": 1.0327, + "step": 505 + }, + { + "epoch": 2.130526315789474, + "grad_norm": 0.392578125, + "learning_rate": 9.096935586929392e-05, + "loss": 0.9649, + "step": 506 + }, + { + "epoch": 2.134736842105263, + "grad_norm": 0.63671875, + "learning_rate": 9.063585104656493e-05, + "loss": 0.6553, + "step": 507 + }, + { + "epoch": 2.1389473684210527, + "grad_norm": 0.4765625, + "learning_rate": 9.030245126559262e-05, + "loss": 0.6973, + "step": 508 + }, + { + "epoch": 2.143157894736842, + "grad_norm": 0.48828125, + "learning_rate": 8.996916026626812e-05, + "loss": 0.7395, + "step": 509 + }, + { + "epoch": 2.1473684210526316, + "grad_norm": 0.70703125, + "learning_rate": 8.963598178726254e-05, + "loss": 1.0481, + "step": 510 + }, + { + "epoch": 2.1515789473684213, + "grad_norm": 0.80859375, + "learning_rate": 8.930291956598462e-05, + "loss": 0.9903, + "step": 511 + }, + { + "epoch": 2.1557894736842105, + "grad_norm": 0.451171875, + "learning_rate": 8.896997733853903e-05, + "loss": 1.2016, + "step": 512 + }, + { + "epoch": 2.16, + "grad_norm": 0.3828125, + "learning_rate": 8.863715883968451e-05, + "loss": 1.2436, + "step": 513 + }, + { + "epoch": 2.1642105263157894, + "grad_norm": 0.333984375, + "learning_rate": 8.830446780279176e-05, + "loss": 0.9862, + "step": 514 + }, + { + "epoch": 2.168421052631579, + "grad_norm": 0.5078125, + "learning_rate": 8.797190795980172e-05, + "loss": 0.5605, + "step": 515 + }, + { + "epoch": 2.1726315789473682, + "grad_norm": 0.6484375, + "learning_rate": 8.763948304118368e-05, + "loss": 0.4958, + "step": 516 + }, + { + "epoch": 2.176842105263158, + "grad_norm": 0.5625, + "learning_rate": 8.730719677589344e-05, + "loss": 1.0046, + "step": 517 + }, + { + "epoch": 2.1810526315789476, + "grad_norm": 0.486328125, + "learning_rate": 8.697505289133145e-05, + "loss": 0.6935, + "step": 518 + }, + { + "epoch": 2.185263157894737, + "grad_norm": 0.37890625, + "learning_rate": 8.6643055113301e-05, + "loss": 1.2277, + "step": 519 + }, + { + "epoch": 2.1894736842105265, + "grad_norm": 0.73828125, + "learning_rate": 8.631120716596641e-05, + "loss": 0.7949, + "step": 520 + }, + { + "epoch": 2.1936842105263157, + "grad_norm": 0.6640625, + "learning_rate": 8.597951277181142e-05, + "loss": 0.8309, + "step": 521 + }, + { + "epoch": 2.1978947368421053, + "grad_norm": 0.58984375, + "learning_rate": 8.564797565159714e-05, + "loss": 0.6418, + "step": 522 + }, + { + "epoch": 2.2021052631578946, + "grad_norm": 0.77734375, + "learning_rate": 8.531659952432056e-05, + "loss": 1.1851, + "step": 523 + }, + { + "epoch": 2.2063157894736842, + "grad_norm": 0.380859375, + "learning_rate": 8.498538810717267e-05, + "loss": 1.015, + "step": 524 + }, + { + "epoch": 2.2105263157894735, + "grad_norm": 0.384765625, + "learning_rate": 8.465434511549691e-05, + "loss": 1.0365, + "step": 525 + }, + { + "epoch": 2.214736842105263, + "grad_norm": 0.41015625, + "learning_rate": 8.432347426274739e-05, + "loss": 1.0889, + "step": 526 + }, + { + "epoch": 2.218947368421053, + "grad_norm": 0.7578125, + "learning_rate": 8.399277926044724e-05, + "loss": 1.0203, + "step": 527 + }, + { + "epoch": 2.223157894736842, + "grad_norm": 0.55078125, + "learning_rate": 8.366226381814697e-05, + "loss": 0.9986, + "step": 528 + }, + { + "epoch": 2.2273684210526317, + "grad_norm": 0.41015625, + "learning_rate": 8.333193164338304e-05, + "loss": 1.1371, + "step": 529 + }, + { + "epoch": 2.231578947368421, + "grad_norm": 0.70703125, + "learning_rate": 8.300178644163594e-05, + "loss": 1.1068, + "step": 530 + }, + { + "epoch": 2.2357894736842105, + "grad_norm": 0.8046875, + "learning_rate": 8.267183191628889e-05, + "loss": 1.0577, + "step": 531 + }, + { + "epoch": 2.24, + "grad_norm": 0.75, + "learning_rate": 8.234207176858614e-05, + "loss": 0.9573, + "step": 532 + }, + { + "epoch": 2.2442105263157894, + "grad_norm": 0.400390625, + "learning_rate": 8.201250969759168e-05, + "loss": 1.4236, + "step": 533 + }, + { + "epoch": 2.248421052631579, + "grad_norm": 0.61328125, + "learning_rate": 8.16831494001474e-05, + "loss": 0.5257, + "step": 534 + }, + { + "epoch": 2.2526315789473683, + "grad_norm": 0.515625, + "learning_rate": 8.13539945708319e-05, + "loss": 1.2213, + "step": 535 + }, + { + "epoch": 2.256842105263158, + "grad_norm": 0.6640625, + "learning_rate": 8.102504890191892e-05, + "loss": 0.8878, + "step": 536 + }, + { + "epoch": 2.261052631578947, + "grad_norm": 0.435546875, + "learning_rate": 8.069631608333605e-05, + "loss": 0.8702, + "step": 537 + }, + { + "epoch": 2.265263157894737, + "grad_norm": 0.482421875, + "learning_rate": 8.036779980262311e-05, + "loss": 0.8551, + "step": 538 + }, + { + "epoch": 2.269473684210526, + "grad_norm": 0.5234375, + "learning_rate": 8.003950374489101e-05, + "loss": 0.6752, + "step": 539 + }, + { + "epoch": 2.2736842105263158, + "grad_norm": 0.546875, + "learning_rate": 7.971143159278026e-05, + "loss": 1.0084, + "step": 540 + }, + { + "epoch": 2.2736842105263158, + "eval_loss": 1.2140343189239502, + "eval_runtime": 25.8753, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.932, + "step": 540 + }, + { + "epoch": 2.2778947368421054, + "grad_norm": 0.349609375, + "learning_rate": 7.938358702641979e-05, + "loss": 1.091, + "step": 541 + }, + { + "epoch": 2.2821052631578946, + "grad_norm": 0.375, + "learning_rate": 7.905597372338558e-05, + "loss": 0.8771, + "step": 542 + }, + { + "epoch": 2.2863157894736843, + "grad_norm": 0.75, + "learning_rate": 7.872859535865935e-05, + "loss": 1.0554, + "step": 543 + }, + { + "epoch": 2.2905263157894735, + "grad_norm": 0.4140625, + "learning_rate": 7.840145560458756e-05, + "loss": 0.958, + "step": 544 + }, + { + "epoch": 2.294736842105263, + "grad_norm": 0.4375, + "learning_rate": 7.807455813083996e-05, + "loss": 0.9956, + "step": 545 + }, + { + "epoch": 2.298947368421053, + "grad_norm": 0.50390625, + "learning_rate": 7.774790660436858e-05, + "loss": 0.864, + "step": 546 + }, + { + "epoch": 2.303157894736842, + "grad_norm": 0.71875, + "learning_rate": 7.742150468936647e-05, + "loss": 0.7358, + "step": 547 + }, + { + "epoch": 2.3073684210526317, + "grad_norm": 0.380859375, + "learning_rate": 7.709535604722684e-05, + "loss": 1.0316, + "step": 548 + }, + { + "epoch": 2.311578947368421, + "grad_norm": 0.64453125, + "learning_rate": 7.67694643365017e-05, + "loss": 0.9351, + "step": 549 + }, + { + "epoch": 2.3157894736842106, + "grad_norm": 0.498046875, + "learning_rate": 7.644383321286094e-05, + "loss": 1.0617, + "step": 550 + }, + { + "epoch": 2.32, + "grad_norm": 0.9140625, + "learning_rate": 7.61184663290514e-05, + "loss": 0.6994, + "step": 551 + }, + { + "epoch": 2.3242105263157895, + "grad_norm": 0.375, + "learning_rate": 7.579336733485584e-05, + "loss": 1.0248, + "step": 552 + }, + { + "epoch": 2.3284210526315787, + "grad_norm": 0.58984375, + "learning_rate": 7.546853987705191e-05, + "loss": 0.9696, + "step": 553 + }, + { + "epoch": 2.3326315789473684, + "grad_norm": 0.7265625, + "learning_rate": 7.514398759937135e-05, + "loss": 0.7576, + "step": 554 + }, + { + "epoch": 2.336842105263158, + "grad_norm": 0.369140625, + "learning_rate": 7.481971414245907e-05, + "loss": 0.7026, + "step": 555 + }, + { + "epoch": 2.3410526315789473, + "grad_norm": 0.69921875, + "learning_rate": 7.449572314383237e-05, + "loss": 0.8625, + "step": 556 + }, + { + "epoch": 2.345263157894737, + "grad_norm": 0.921875, + "learning_rate": 7.417201823784005e-05, + "loss": 0.849, + "step": 557 + }, + { + "epoch": 2.349473684210526, + "grad_norm": 0.453125, + "learning_rate": 7.384860305562172e-05, + "loss": 1.274, + "step": 558 + }, + { + "epoch": 2.353684210526316, + "grad_norm": 0.73828125, + "learning_rate": 7.35254812250669e-05, + "loss": 1.0279, + "step": 559 + }, + { + "epoch": 2.3578947368421055, + "grad_norm": 0.353515625, + "learning_rate": 7.320265637077473e-05, + "loss": 1.0411, + "step": 560 + }, + { + "epoch": 2.3621052631578947, + "grad_norm": 0.7109375, + "learning_rate": 7.288013211401278e-05, + "loss": 1.1542, + "step": 561 + }, + { + "epoch": 2.3663157894736844, + "grad_norm": 0.55078125, + "learning_rate": 7.255791207267679e-05, + "loss": 1.1941, + "step": 562 + }, + { + "epoch": 2.3705263157894736, + "grad_norm": 0.58203125, + "learning_rate": 7.223599986124994e-05, + "loss": 1.0699, + "step": 563 + }, + { + "epoch": 2.3747368421052633, + "grad_norm": 0.76953125, + "learning_rate": 7.191439909076243e-05, + "loss": 1.0074, + "step": 564 + }, + { + "epoch": 2.3789473684210525, + "grad_norm": 0.408203125, + "learning_rate": 7.159311336875083e-05, + "loss": 1.1968, + "step": 565 + }, + { + "epoch": 2.383157894736842, + "grad_norm": 1.0234375, + "learning_rate": 7.127214629921765e-05, + "loss": 0.9567, + "step": 566 + }, + { + "epoch": 2.3873684210526314, + "grad_norm": 0.5, + "learning_rate": 7.095150148259093e-05, + "loss": 0.585, + "step": 567 + }, + { + "epoch": 2.391578947368421, + "grad_norm": 0.52734375, + "learning_rate": 7.0631182515684e-05, + "loss": 0.9657, + "step": 568 + }, + { + "epoch": 2.3957894736842107, + "grad_norm": 0.43359375, + "learning_rate": 7.031119299165483e-05, + "loss": 0.7711, + "step": 569 + }, + { + "epoch": 2.4, + "grad_norm": 0.43359375, + "learning_rate": 6.999153649996595e-05, + "loss": 0.9455, + "step": 570 + }, + { + "epoch": 2.4042105263157896, + "grad_norm": 0.6953125, + "learning_rate": 6.967221662634407e-05, + "loss": 0.9143, + "step": 571 + }, + { + "epoch": 2.408421052631579, + "grad_norm": 0.443359375, + "learning_rate": 6.935323695274002e-05, + "loss": 1.3066, + "step": 572 + }, + { + "epoch": 2.4126315789473685, + "grad_norm": 0.4140625, + "learning_rate": 6.903460105728834e-05, + "loss": 1.02, + "step": 573 + }, + { + "epoch": 2.416842105263158, + "grad_norm": 0.322265625, + "learning_rate": 6.871631251426728e-05, + "loss": 1.0387, + "step": 574 + }, + { + "epoch": 2.4210526315789473, + "grad_norm": 0.625, + "learning_rate": 6.839837489405881e-05, + "loss": 0.9289, + "step": 575 + }, + { + "epoch": 2.425263157894737, + "grad_norm": 0.34765625, + "learning_rate": 6.808079176310827e-05, + "loss": 1.0434, + "step": 576 + }, + { + "epoch": 2.4294736842105262, + "grad_norm": 0.69921875, + "learning_rate": 6.776356668388464e-05, + "loss": 0.9514, + "step": 577 + }, + { + "epoch": 2.433684210526316, + "grad_norm": 0.486328125, + "learning_rate": 6.744670321484043e-05, + "loss": 0.8505, + "step": 578 + }, + { + "epoch": 2.437894736842105, + "grad_norm": 0.447265625, + "learning_rate": 6.713020491037188e-05, + "loss": 1.2355, + "step": 579 + }, + { + "epoch": 2.442105263157895, + "grad_norm": 0.359375, + "learning_rate": 6.681407532077895e-05, + "loss": 0.9406, + "step": 580 + }, + { + "epoch": 2.446315789473684, + "grad_norm": 0.6953125, + "learning_rate": 6.649831799222557e-05, + "loss": 0.8115, + "step": 581 + }, + { + "epoch": 2.4505263157894737, + "grad_norm": 0.5390625, + "learning_rate": 6.618293646669986e-05, + "loss": 1.0947, + "step": 582 + }, + { + "epoch": 2.4547368421052633, + "grad_norm": 0.69921875, + "learning_rate": 6.586793428197447e-05, + "loss": 1.1178, + "step": 583 + }, + { + "epoch": 2.4589473684210525, + "grad_norm": 0.427734375, + "learning_rate": 6.555331497156672e-05, + "loss": 0.8136, + "step": 584 + }, + { + "epoch": 2.463157894736842, + "grad_norm": 0.87109375, + "learning_rate": 6.523908206469909e-05, + "loss": 1.0041, + "step": 585 + }, + { + "epoch": 2.4673684210526314, + "grad_norm": 0.4375, + "learning_rate": 6.492523908625959e-05, + "loss": 1.1051, + "step": 586 + }, + { + "epoch": 2.471578947368421, + "grad_norm": 0.5625, + "learning_rate": 6.461178955676228e-05, + "loss": 1.0417, + "step": 587 + }, + { + "epoch": 2.4757894736842108, + "grad_norm": 0.35546875, + "learning_rate": 6.42987369923077e-05, + "loss": 0.9521, + "step": 588 + }, + { + "epoch": 2.48, + "grad_norm": 0.375, + "learning_rate": 6.398608490454346e-05, + "loss": 1.3175, + "step": 589 + }, + { + "epoch": 2.4842105263157896, + "grad_norm": 0.40625, + "learning_rate": 6.367383680062478e-05, + "loss": 1.0413, + "step": 590 + }, + { + "epoch": 2.488421052631579, + "grad_norm": 0.625, + "learning_rate": 6.336199618317537e-05, + "loss": 0.8431, + "step": 591 + }, + { + "epoch": 2.4926315789473685, + "grad_norm": 0.451171875, + "learning_rate": 6.30505665502479e-05, + "loss": 0.8335, + "step": 592 + }, + { + "epoch": 2.4968421052631578, + "grad_norm": 0.83203125, + "learning_rate": 6.27395513952848e-05, + "loss": 0.689, + "step": 593 + }, + { + "epoch": 2.5010526315789474, + "grad_norm": 0.3984375, + "learning_rate": 6.242895420707917e-05, + "loss": 1.1857, + "step": 594 + }, + { + "epoch": 2.5052631578947366, + "grad_norm": 0.375, + "learning_rate": 6.211877846973563e-05, + "loss": 0.6606, + "step": 595 + }, + { + "epoch": 2.5094736842105263, + "grad_norm": 0.578125, + "learning_rate": 6.180902766263113e-05, + "loss": 0.892, + "step": 596 + }, + { + "epoch": 2.513684210526316, + "grad_norm": 0.427734375, + "learning_rate": 6.149970526037603e-05, + "loss": 1.0731, + "step": 597 + }, + { + "epoch": 2.517894736842105, + "grad_norm": 0.578125, + "learning_rate": 6.119081473277501e-05, + "loss": 1.0982, + "step": 598 + }, + { + "epoch": 2.522105263157895, + "grad_norm": 0.376953125, + "learning_rate": 6.088235954478842e-05, + "loss": 1.0339, + "step": 599 + }, + { + "epoch": 2.526315789473684, + "grad_norm": 0.384765625, + "learning_rate": 6.057434315649304e-05, + "loss": 1.1655, + "step": 600 + }, + { + "epoch": 2.526315789473684, + "eval_loss": 1.217448353767395, + "eval_runtime": 25.9078, + "eval_samples_per_second": 3.86, + "eval_steps_per_second": 1.93, + "step": 600 + }, + { + "epoch": 2.5305263157894737, + "grad_norm": 0.953125, + "learning_rate": 6.026676902304345e-05, + "loss": 0.8561, + "step": 601 + }, + { + "epoch": 2.5347368421052634, + "grad_norm": 0.625, + "learning_rate": 5.99596405946333e-05, + "loss": 0.8852, + "step": 602 + }, + { + "epoch": 2.5389473684210526, + "grad_norm": 0.423828125, + "learning_rate": 5.965296131645663e-05, + "loss": 0.8336, + "step": 603 + }, + { + "epoch": 2.543157894736842, + "grad_norm": 1.3515625, + "learning_rate": 5.9346734628669065e-05, + "loss": 0.6848, + "step": 604 + }, + { + "epoch": 2.5473684210526315, + "grad_norm": 0.8046875, + "learning_rate": 5.904096396634935e-05, + "loss": 0.951, + "step": 605 + }, + { + "epoch": 2.551578947368421, + "grad_norm": 0.44921875, + "learning_rate": 5.873565275946088e-05, + "loss": 0.877, + "step": 606 + }, + { + "epoch": 2.5557894736842104, + "grad_norm": 0.60546875, + "learning_rate": 5.843080443281303e-05, + "loss": 0.9183, + "step": 607 + }, + { + "epoch": 2.56, + "grad_norm": 0.71484375, + "learning_rate": 5.8126422406022885e-05, + "loss": 0.6659, + "step": 608 + }, + { + "epoch": 2.5642105263157893, + "grad_norm": 0.54296875, + "learning_rate": 5.782251009347679e-05, + "loss": 1.0883, + "step": 609 + }, + { + "epoch": 2.568421052631579, + "grad_norm": 0.453125, + "learning_rate": 5.7519070904292247e-05, + "loss": 0.963, + "step": 610 + }, + { + "epoch": 2.5726315789473686, + "grad_norm": 0.43359375, + "learning_rate": 5.7216108242279386e-05, + "loss": 1.2451, + "step": 611 + }, + { + "epoch": 2.576842105263158, + "grad_norm": 0.78515625, + "learning_rate": 5.691362550590297e-05, + "loss": 0.8804, + "step": 612 + }, + { + "epoch": 2.5810526315789475, + "grad_norm": 0.8515625, + "learning_rate": 5.6611626088244194e-05, + "loss": 1.0411, + "step": 613 + }, + { + "epoch": 2.5852631578947367, + "grad_norm": 0.42578125, + "learning_rate": 5.631011337696271e-05, + "loss": 0.9097, + "step": 614 + }, + { + "epoch": 2.5894736842105264, + "grad_norm": 0.478515625, + "learning_rate": 5.6009090754258545e-05, + "loss": 1.0427, + "step": 615 + }, + { + "epoch": 2.593684210526316, + "grad_norm": 0.40625, + "learning_rate": 5.570856159683418e-05, + "loss": 0.8898, + "step": 616 + }, + { + "epoch": 2.5978947368421053, + "grad_norm": 0.373046875, + "learning_rate": 5.5408529275856624e-05, + "loss": 0.9602, + "step": 617 + }, + { + "epoch": 2.6021052631578945, + "grad_norm": 0.5390625, + "learning_rate": 5.510899715691984e-05, + "loss": 1.0589, + "step": 618 + }, + { + "epoch": 2.606315789473684, + "grad_norm": 0.4765625, + "learning_rate": 5.4809968600006635e-05, + "loss": 0.939, + "step": 619 + }, + { + "epoch": 2.610526315789474, + "grad_norm": 0.44921875, + "learning_rate": 5.451144695945116e-05, + "loss": 0.909, + "step": 620 + }, + { + "epoch": 2.614736842105263, + "grad_norm": 1.203125, + "learning_rate": 5.421343558390128e-05, + "loss": 1.0197, + "step": 621 + }, + { + "epoch": 2.6189473684210527, + "grad_norm": 0.96875, + "learning_rate": 5.3915937816281095e-05, + "loss": 0.991, + "step": 622 + }, + { + "epoch": 2.623157894736842, + "grad_norm": 0.85546875, + "learning_rate": 5.361895699375319e-05, + "loss": 0.9968, + "step": 623 + }, + { + "epoch": 2.6273684210526316, + "grad_norm": 0.890625, + "learning_rate": 5.3322496447681414e-05, + "loss": 1.0106, + "step": 624 + }, + { + "epoch": 2.6315789473684212, + "grad_norm": 0.41015625, + "learning_rate": 5.302655950359341e-05, + "loss": 1.0636, + "step": 625 + }, + { + "epoch": 2.6357894736842105, + "grad_norm": 0.44921875, + "learning_rate": 5.273114948114346e-05, + "loss": 0.7656, + "step": 626 + }, + { + "epoch": 2.64, + "grad_norm": 0.5078125, + "learning_rate": 5.2436269694075046e-05, + "loss": 1.1255, + "step": 627 + }, + { + "epoch": 2.6442105263157893, + "grad_norm": 2.515625, + "learning_rate": 5.214192345018374e-05, + "loss": 0.8405, + "step": 628 + }, + { + "epoch": 2.648421052631579, + "grad_norm": 0.375, + "learning_rate": 5.1848114051280195e-05, + "loss": 1.0719, + "step": 629 + }, + { + "epoch": 2.6526315789473687, + "grad_norm": 0.353515625, + "learning_rate": 5.1554844793153e-05, + "loss": 0.9582, + "step": 630 + }, + { + "epoch": 2.656842105263158, + "grad_norm": 0.46875, + "learning_rate": 5.126211896553178e-05, + "loss": 0.8587, + "step": 631 + }, + { + "epoch": 2.661052631578947, + "grad_norm": 0.8203125, + "learning_rate": 5.096993985205023e-05, + "loss": 1.055, + "step": 632 + }, + { + "epoch": 2.665263157894737, + "grad_norm": 0.48046875, + "learning_rate": 5.0678310730209275e-05, + "loss": 1.0103, + "step": 633 + }, + { + "epoch": 2.6694736842105264, + "grad_norm": 0.70703125, + "learning_rate": 5.0387234871340486e-05, + "loss": 0.7718, + "step": 634 + }, + { + "epoch": 2.6736842105263157, + "grad_norm": 0.72265625, + "learning_rate": 5.009671554056911e-05, + "loss": 0.9338, + "step": 635 + }, + { + "epoch": 2.6778947368421053, + "grad_norm": 0.373046875, + "learning_rate": 4.980675599677757e-05, + "loss": 0.7023, + "step": 636 + }, + { + "epoch": 2.6821052631578945, + "grad_norm": 0.66796875, + "learning_rate": 4.9517359492569035e-05, + "loss": 1.1044, + "step": 637 + }, + { + "epoch": 2.686315789473684, + "grad_norm": 0.5234375, + "learning_rate": 4.9228529274230695e-05, + "loss": 0.9719, + "step": 638 + }, + { + "epoch": 2.690526315789474, + "grad_norm": 0.55078125, + "learning_rate": 4.894026858169752e-05, + "loss": 1.1118, + "step": 639 + }, + { + "epoch": 2.694736842105263, + "grad_norm": 0.490234375, + "learning_rate": 4.865258064851579e-05, + "loss": 1.2393, + "step": 640 + }, + { + "epoch": 2.6989473684210528, + "grad_norm": 0.443359375, + "learning_rate": 4.8365468701807016e-05, + "loss": 0.8362, + "step": 641 + }, + { + "epoch": 2.703157894736842, + "grad_norm": 0.82421875, + "learning_rate": 4.807893596223152e-05, + "loss": 0.9748, + "step": 642 + }, + { + "epoch": 2.7073684210526316, + "grad_norm": 0.6328125, + "learning_rate": 4.7792985643952435e-05, + "loss": 1.1939, + "step": 643 + }, + { + "epoch": 2.7115789473684213, + "grad_norm": 0.51171875, + "learning_rate": 4.75076209545996e-05, + "loss": 0.8513, + "step": 644 + }, + { + "epoch": 2.7157894736842105, + "grad_norm": 0.47265625, + "learning_rate": 4.722284509523362e-05, + "loss": 0.8377, + "step": 645 + }, + { + "epoch": 2.7199999999999998, + "grad_norm": 0.7109375, + "learning_rate": 4.693866126030995e-05, + "loss": 1.0126, + "step": 646 + }, + { + "epoch": 2.7242105263157894, + "grad_norm": 0.57421875, + "learning_rate": 4.665507263764299e-05, + "loss": 0.6115, + "step": 647 + }, + { + "epoch": 2.728421052631579, + "grad_norm": 0.88671875, + "learning_rate": 4.637208240837042e-05, + "loss": 1.0717, + "step": 648 + }, + { + "epoch": 2.7326315789473683, + "grad_norm": 0.56640625, + "learning_rate": 4.608969374691753e-05, + "loss": 1.1582, + "step": 649 + }, + { + "epoch": 2.736842105263158, + "grad_norm": 0.91796875, + "learning_rate": 4.5807909820961494e-05, + "loss": 1.1218, + "step": 650 + }, + { + "epoch": 2.741052631578947, + "grad_norm": 0.6328125, + "learning_rate": 4.5526733791395896e-05, + "loss": 0.9114, + "step": 651 + }, + { + "epoch": 2.745263157894737, + "grad_norm": 0.70703125, + "learning_rate": 4.5246168812295286e-05, + "loss": 0.957, + "step": 652 + }, + { + "epoch": 2.7494736842105265, + "grad_norm": 0.5859375, + "learning_rate": 4.496621803087988e-05, + "loss": 0.6728, + "step": 653 + }, + { + "epoch": 2.7536842105263157, + "grad_norm": 1.2109375, + "learning_rate": 4.468688458748006e-05, + "loss": 0.8425, + "step": 654 + }, + { + "epoch": 2.7578947368421054, + "grad_norm": 0.5703125, + "learning_rate": 4.4408171615501316e-05, + "loss": 0.6478, + "step": 655 + }, + { + "epoch": 2.7621052631578946, + "grad_norm": 0.28125, + "learning_rate": 4.413008224138897e-05, + "loss": 0.9384, + "step": 656 + }, + { + "epoch": 2.7663157894736843, + "grad_norm": 0.734375, + "learning_rate": 4.38526195845933e-05, + "loss": 0.9842, + "step": 657 + }, + { + "epoch": 2.770526315789474, + "grad_norm": 0.8125, + "learning_rate": 4.357578675753432e-05, + "loss": 0.8117, + "step": 658 + }, + { + "epoch": 2.774736842105263, + "grad_norm": 0.6328125, + "learning_rate": 4.329958686556699e-05, + "loss": 0.8619, + "step": 659 + }, + { + "epoch": 2.7789473684210524, + "grad_norm": 0.55859375, + "learning_rate": 4.302402300694636e-05, + "loss": 1.1503, + "step": 660 + }, + { + "epoch": 2.7789473684210524, + "eval_loss": 1.2197896242141724, + "eval_runtime": 25.8587, + "eval_samples_per_second": 3.867, + "eval_steps_per_second": 1.934, + "step": 660 + }, + { + "epoch": 2.783157894736842, + "grad_norm": 0.59765625, + "learning_rate": 4.274909827279283e-05, + "loss": 0.7691, + "step": 661 + }, + { + "epoch": 2.7873684210526317, + "grad_norm": 0.54296875, + "learning_rate": 4.247481574705744e-05, + "loss": 0.5837, + "step": 662 + }, + { + "epoch": 2.791578947368421, + "grad_norm": 0.80078125, + "learning_rate": 4.220117850648733e-05, + "loss": 0.8718, + "step": 663 + }, + { + "epoch": 2.7957894736842106, + "grad_norm": 0.75390625, + "learning_rate": 4.1928189620591116e-05, + "loss": 0.8497, + "step": 664 + }, + { + "epoch": 2.8, + "grad_norm": 0.359375, + "learning_rate": 4.165585215160468e-05, + "loss": 1.1803, + "step": 665 + }, + { + "epoch": 2.8042105263157895, + "grad_norm": 0.78515625, + "learning_rate": 4.138416915445655e-05, + "loss": 0.7498, + "step": 666 + }, + { + "epoch": 2.808421052631579, + "grad_norm": 0.451171875, + "learning_rate": 4.1113143676733757e-05, + "loss": 1.2798, + "step": 667 + }, + { + "epoch": 2.8126315789473684, + "grad_norm": 0.341796875, + "learning_rate": 4.084277875864776e-05, + "loss": 0.9448, + "step": 668 + }, + { + "epoch": 2.816842105263158, + "grad_norm": 0.625, + "learning_rate": 4.0573077433000115e-05, + "loss": 1.0791, + "step": 669 + }, + { + "epoch": 2.8210526315789473, + "grad_norm": 0.3671875, + "learning_rate": 4.030404272514864e-05, + "loss": 0.9928, + "step": 670 + }, + { + "epoch": 2.825263157894737, + "grad_norm": 0.6015625, + "learning_rate": 4.003567765297331e-05, + "loss": 1.0038, + "step": 671 + }, + { + "epoch": 2.8294736842105266, + "grad_norm": 0.46875, + "learning_rate": 3.9767985226842696e-05, + "loss": 1.294, + "step": 672 + }, + { + "epoch": 2.833684210526316, + "grad_norm": 0.34765625, + "learning_rate": 3.950096844957981e-05, + "loss": 1.1689, + "step": 673 + }, + { + "epoch": 2.837894736842105, + "grad_norm": 0.609375, + "learning_rate": 3.923463031642872e-05, + "loss": 1.2769, + "step": 674 + }, + { + "epoch": 2.8421052631578947, + "grad_norm": 0.43359375, + "learning_rate": 3.8968973815020806e-05, + "loss": 1.0042, + "step": 675 + }, + { + "epoch": 2.8463157894736844, + "grad_norm": 0.58984375, + "learning_rate": 3.870400192534128e-05, + "loss": 0.8091, + "step": 676 + }, + { + "epoch": 2.8505263157894736, + "grad_norm": 0.5, + "learning_rate": 3.843971761969583e-05, + "loss": 0.9447, + "step": 677 + }, + { + "epoch": 2.8547368421052632, + "grad_norm": 0.58984375, + "learning_rate": 3.81761238626771e-05, + "loss": 0.7497, + "step": 678 + }, + { + "epoch": 2.8589473684210525, + "grad_norm": 0.66015625, + "learning_rate": 3.791322361113164e-05, + "loss": 0.7618, + "step": 679 + }, + { + "epoch": 2.863157894736842, + "grad_norm": 0.5078125, + "learning_rate": 3.7651019814126654e-05, + "loss": 1.1192, + "step": 680 + }, + { + "epoch": 2.867368421052632, + "grad_norm": 0.484375, + "learning_rate": 3.738951541291685e-05, + "loss": 0.8821, + "step": 681 + }, + { + "epoch": 2.871578947368421, + "grad_norm": 0.625, + "learning_rate": 3.7128713340911535e-05, + "loss": 0.868, + "step": 682 + }, + { + "epoch": 2.8757894736842107, + "grad_norm": 0.328125, + "learning_rate": 3.686861652364164e-05, + "loss": 1.0257, + "step": 683 + }, + { + "epoch": 2.88, + "grad_norm": 0.376953125, + "learning_rate": 3.660922787872706e-05, + "loss": 0.8358, + "step": 684 + }, + { + "epoch": 2.8842105263157896, + "grad_norm": 0.470703125, + "learning_rate": 3.635055031584369e-05, + "loss": 1.0488, + "step": 685 + }, + { + "epoch": 2.8884210526315792, + "grad_norm": 0.4140625, + "learning_rate": 3.609258673669097e-05, + "loss": 1.0167, + "step": 686 + }, + { + "epoch": 2.8926315789473684, + "grad_norm": 0.44140625, + "learning_rate": 3.58353400349592e-05, + "loss": 1.1384, + "step": 687 + }, + { + "epoch": 2.8968421052631577, + "grad_norm": 0.69921875, + "learning_rate": 3.557881309629729e-05, + "loss": 0.847, + "step": 688 + }, + { + "epoch": 2.9010526315789473, + "grad_norm": 1.0703125, + "learning_rate": 3.532300879828013e-05, + "loss": 0.9456, + "step": 689 + }, + { + "epoch": 2.905263157894737, + "grad_norm": 0.890625, + "learning_rate": 3.5067930010376484e-05, + "loss": 0.8495, + "step": 690 + }, + { + "epoch": 2.909473684210526, + "grad_norm": 0.91015625, + "learning_rate": 3.481357959391671e-05, + "loss": 1.0274, + "step": 691 + }, + { + "epoch": 2.913684210526316, + "grad_norm": 0.4765625, + "learning_rate": 3.455996040206076e-05, + "loss": 1.0327, + "step": 692 + }, + { + "epoch": 2.917894736842105, + "grad_norm": 0.64453125, + "learning_rate": 3.43070752797661e-05, + "loss": 0.83, + "step": 693 + }, + { + "epoch": 2.9221052631578948, + "grad_norm": 0.8671875, + "learning_rate": 3.4054927063755796e-05, + "loss": 0.786, + "step": 694 + }, + { + "epoch": 2.9263157894736844, + "grad_norm": 0.67578125, + "learning_rate": 3.38035185824867e-05, + "loss": 0.8956, + "step": 695 + }, + { + "epoch": 2.9305263157894736, + "grad_norm": 0.61328125, + "learning_rate": 3.355285265611784e-05, + "loss": 0.8357, + "step": 696 + }, + { + "epoch": 2.9347368421052633, + "grad_norm": 0.498046875, + "learning_rate": 3.3302932096478554e-05, + "loss": 1.0673, + "step": 697 + }, + { + "epoch": 2.9389473684210525, + "grad_norm": 0.53515625, + "learning_rate": 3.305375970703711e-05, + "loss": 0.8016, + "step": 698 + }, + { + "epoch": 2.943157894736842, + "grad_norm": 0.62890625, + "learning_rate": 3.28053382828693e-05, + "loss": 0.817, + "step": 699 + }, + { + "epoch": 2.9473684210526314, + "grad_norm": 0.62890625, + "learning_rate": 3.2557670610626925e-05, + "loss": 1.0499, + "step": 700 + }, + { + "epoch": 2.951578947368421, + "grad_norm": 0.9375, + "learning_rate": 3.231075946850663e-05, + "loss": 0.8511, + "step": 701 + }, + { + "epoch": 2.9557894736842103, + "grad_norm": 0.455078125, + "learning_rate": 3.206460762621873e-05, + "loss": 1.0486, + "step": 702 + }, + { + "epoch": 2.96, + "grad_norm": 0.38671875, + "learning_rate": 3.1819217844956214e-05, + "loss": 0.9743, + "step": 703 + }, + { + "epoch": 2.9642105263157896, + "grad_norm": 0.42578125, + "learning_rate": 3.157459287736362e-05, + "loss": 0.9247, + "step": 704 + }, + { + "epoch": 2.968421052631579, + "grad_norm": 0.3359375, + "learning_rate": 3.1330735467506246e-05, + "loss": 1.0188, + "step": 705 + }, + { + "epoch": 2.9726315789473685, + "grad_norm": 0.51953125, + "learning_rate": 3.108764835083938e-05, + "loss": 1.1879, + "step": 706 + }, + { + "epoch": 2.9768421052631577, + "grad_norm": 0.41015625, + "learning_rate": 3.084533425417758e-05, + "loss": 0.6776, + "step": 707 + }, + { + "epoch": 2.9810526315789474, + "grad_norm": 0.349609375, + "learning_rate": 3.0603795895664124e-05, + "loss": 0.9761, + "step": 708 + }, + { + "epoch": 2.985263157894737, + "grad_norm": 0.7109375, + "learning_rate": 3.036303598474045e-05, + "loss": 0.9003, + "step": 709 + }, + { + "epoch": 2.9894736842105263, + "grad_norm": 0.38671875, + "learning_rate": 3.0123057222115836e-05, + "loss": 1.0767, + "step": 710 + }, + { + "epoch": 2.993684210526316, + "grad_norm": 0.52734375, + "learning_rate": 2.988386229973713e-05, + "loss": 0.792, + "step": 711 + } + ], + "logging_steps": 1, + "max_steps": 948, + "num_input_tokens_seen": 0, + "num_train_epochs": 4, + "save_steps": 237, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.1832233894608896e+16, + "train_batch_size": 2, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-711/training_args.bin b/checkpoint-711/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..bd7f9976bbc503fbce44b538eec2dade4024c02a --- /dev/null +++ b/checkpoint-711/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee017c711219d88132625f3e50d1a67d1c6236a55d3f5a2ad9f499310d968cef +size 5944 diff --git a/checkpoint-948/README.md b/checkpoint-948/README.md new file mode 100644 index 0000000000000000000000000000000000000000..e1ccd431539a8f1507d8755a9c3ba5e5b2897978 --- /dev/null +++ b/checkpoint-948/README.md @@ -0,0 +1,202 @@ +--- +library_name: peft +base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **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] +### Framework versions + +- PEFT 0.11.1 \ No newline at end of file diff --git a/checkpoint-948/adapter_config.json b/checkpoint-948/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..cfa45a23cb958fe6ad106d4adf7f6169698c1fc8 --- /dev/null +++ b/checkpoint-948/adapter_config.json @@ -0,0 +1,34 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 16, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "v_proj", + "k_proj", + "q_proj", + "gate_proj", + "up_proj", + "down_proj", + "o_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-948/adapter_model.safetensors b/checkpoint-948/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..39133f39a81d54ae9e499f18ba536cfbea2e801c --- /dev/null +++ b/checkpoint-948/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bbe388d987f9b4a7486827806ff2dbb1a347a4d34f04b03169d6bf3098290369 +size 50503848 diff --git a/checkpoint-948/optimizer.pt b/checkpoint-948/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..b492904665dc3c5104b40431de49e19b39756191 --- /dev/null +++ b/checkpoint-948/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3f7ccbc0992f3b6ba1f71ee713f8f3fb4b6aba77cdd6bd7bf05522d87c22d4ea +size 202035770 diff --git a/checkpoint-948/rng_state.pth b/checkpoint-948/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..d9a533d223b3e816aebc8e01069de080c2fc1466 --- /dev/null +++ b/checkpoint-948/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5da40348f8e17b04ef2d2acf88108eb73604897a344bb8f7a0555a7c1e686a3c +size 14244 diff --git a/checkpoint-948/scheduler.pt b/checkpoint-948/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..81dcf835210966e487c612c1867c7053f09214e6 --- /dev/null +++ b/checkpoint-948/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c026f009280473e2b93182b16ddf0967c143d42c2c00e0cf6d8a798838077f81 +size 1064 diff --git a/checkpoint-948/special_tokens_map.json b/checkpoint-948/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1 --- /dev/null +++ b/checkpoint-948/special_tokens_map.json @@ -0,0 +1,24 @@ +{ + "bos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "", + "unk_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-948/tokenizer.model b/checkpoint-948/tokenizer.model new file mode 100644 index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899 --- /dev/null +++ b/checkpoint-948/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 +size 499723 diff --git a/checkpoint-948/tokenizer_config.json b/checkpoint-948/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..0773857a13ba5a27453a0b462624fe76e8e82a86 --- /dev/null +++ b/checkpoint-948/tokenizer_config.json @@ -0,0 +1,44 @@ +{ + "add_bos_token": true, + "add_eos_token": false, + "add_prefix_space": true, + "added_tokens_decoder": { + "0": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "1": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "2": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "", + "clean_up_tokenization_spaces": false, + "eos_token": "", + "legacy": false, + "model_max_length": 1000000000000000019884624838656, + "pad_token": "", + "padding_side": "right", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "LlamaTokenizer", + "unk_token": "", + "use_default_system_prompt": false, + "use_fast": true +} diff --git a/checkpoint-948/trainer_state.json b/checkpoint-948/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..780189bda2ac0bc4e3ffd5ad62d8d1cc5553a516 --- /dev/null +++ b/checkpoint-948/trainer_state.json @@ -0,0 +1,6797 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 3.991578947368421, + "eval_steps": 60, + "global_step": 948, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.004210526315789474, + "grad_norm": 0.310546875, + "learning_rate": 2e-05, + "loss": 1.5605, + "step": 1 + }, + { + "epoch": 0.004210526315789474, + "eval_loss": 1.5264556407928467, + "eval_runtime": 26.3082, + "eval_samples_per_second": 3.801, + "eval_steps_per_second": 1.901, + "step": 1 + }, + { + "epoch": 0.008421052631578947, + "grad_norm": 0.306640625, + "learning_rate": 4e-05, + "loss": 1.4702, + "step": 2 + }, + { + "epoch": 0.01263157894736842, + "grad_norm": 0.2578125, + "learning_rate": 6e-05, + "loss": 1.4576, + "step": 3 + }, + { + "epoch": 0.016842105263157894, + "grad_norm": 0.65234375, + "learning_rate": 8e-05, + "loss": 1.5868, + "step": 4 + }, + { + "epoch": 0.021052631578947368, + "grad_norm": 0.56640625, + "learning_rate": 0.0001, + "loss": 1.5916, + "step": 5 + }, + { + "epoch": 0.02526315789473684, + "grad_norm": 0.5703125, + "learning_rate": 0.00012, + "loss": 1.7851, + "step": 6 + }, + { + "epoch": 0.029473684210526315, + "grad_norm": 0.3359375, + "learning_rate": 0.00014, + "loss": 1.5319, + "step": 7 + }, + { + "epoch": 0.03368421052631579, + "grad_norm": 0.39453125, + "learning_rate": 0.00016, + "loss": 1.8462, + "step": 8 + }, + { + "epoch": 0.037894736842105266, + "grad_norm": 0.484375, + "learning_rate": 0.00018, + "loss": 1.4219, + "step": 9 + }, + { + "epoch": 0.042105263157894736, + "grad_norm": 0.30859375, + "learning_rate": 0.0002, + "loss": 1.6555, + "step": 10 + }, + { + "epoch": 0.04631578947368421, + "grad_norm": 0.267578125, + "learning_rate": 0.0001999994391281196, + "loss": 1.1527, + "step": 11 + }, + { + "epoch": 0.05052631578947368, + "grad_norm": 0.298828125, + "learning_rate": 0.00019999775651876987, + "loss": 1.4265, + "step": 12 + }, + { + "epoch": 0.05473684210526316, + "grad_norm": 0.376953125, + "learning_rate": 0.00019999495219082545, + "loss": 1.5505, + "step": 13 + }, + { + "epoch": 0.05894736842105263, + "grad_norm": 0.380859375, + "learning_rate": 0.00019999102617574365, + "loss": 1.0313, + "step": 14 + }, + { + "epoch": 0.06315789473684211, + "grad_norm": 0.494140625, + "learning_rate": 0.0001999859785175643, + "loss": 1.439, + "step": 15 + }, + { + "epoch": 0.06736842105263158, + "grad_norm": 0.40234375, + "learning_rate": 0.00019997980927290927, + "loss": 1.3511, + "step": 16 + }, + { + "epoch": 0.07157894736842105, + "grad_norm": 0.390625, + "learning_rate": 0.0001999725185109816, + "loss": 1.2886, + "step": 17 + }, + { + "epoch": 0.07578947368421053, + "grad_norm": 0.388671875, + "learning_rate": 0.00019996410631356498, + "loss": 1.3057, + "step": 18 + }, + { + "epoch": 0.08, + "grad_norm": 0.6328125, + "learning_rate": 0.00019995457277502271, + "loss": 1.7375, + "step": 19 + }, + { + "epoch": 0.08421052631578947, + "grad_norm": 0.51953125, + "learning_rate": 0.00019994391800229666, + "loss": 1.3436, + "step": 20 + }, + { + "epoch": 0.08842105263157894, + "grad_norm": 0.345703125, + "learning_rate": 0.00019993214211490608, + "loss": 1.14, + "step": 21 + }, + { + "epoch": 0.09263157894736843, + "grad_norm": 0.357421875, + "learning_rate": 0.00019991924524494627, + "loss": 1.1322, + "step": 22 + }, + { + "epoch": 0.0968421052631579, + "grad_norm": 0.431640625, + "learning_rate": 0.00019990522753708705, + "loss": 1.4512, + "step": 23 + }, + { + "epoch": 0.10105263157894737, + "grad_norm": 0.2490234375, + "learning_rate": 0.00019989008914857116, + "loss": 1.3419, + "step": 24 + }, + { + "epoch": 0.10526315789473684, + "grad_norm": 0.78515625, + "learning_rate": 0.00019987383024921255, + "loss": 1.382, + "step": 25 + }, + { + "epoch": 0.10947368421052632, + "grad_norm": 0.56640625, + "learning_rate": 0.0001998564510213944, + "loss": 1.582, + "step": 26 + }, + { + "epoch": 0.11368421052631579, + "grad_norm": 0.5703125, + "learning_rate": 0.00019983795166006715, + "loss": 1.188, + "step": 27 + }, + { + "epoch": 0.11789473684210526, + "grad_norm": 0.46875, + "learning_rate": 0.00019981833237274618, + "loss": 1.4246, + "step": 28 + }, + { + "epoch": 0.12210526315789473, + "grad_norm": 0.3984375, + "learning_rate": 0.00019979759337950968, + "loss": 1.2828, + "step": 29 + }, + { + "epoch": 0.12631578947368421, + "grad_norm": 0.55078125, + "learning_rate": 0.00019977573491299598, + "loss": 1.6199, + "step": 30 + }, + { + "epoch": 0.13052631578947368, + "grad_norm": 0.8359375, + "learning_rate": 0.00019975275721840103, + "loss": 1.2398, + "step": 31 + }, + { + "epoch": 0.13473684210526315, + "grad_norm": 0.5, + "learning_rate": 0.00019972866055347572, + "loss": 1.4295, + "step": 32 + }, + { + "epoch": 0.13894736842105262, + "grad_norm": 0.5390625, + "learning_rate": 0.0001997034451885229, + "loss": 1.2814, + "step": 33 + }, + { + "epoch": 0.1431578947368421, + "grad_norm": 0.28515625, + "learning_rate": 0.0001996771114063943, + "loss": 1.532, + "step": 34 + }, + { + "epoch": 0.14736842105263157, + "grad_norm": 0.40625, + "learning_rate": 0.00019964965950248757, + "loss": 1.3516, + "step": 35 + }, + { + "epoch": 0.15157894736842106, + "grad_norm": 0.44921875, + "learning_rate": 0.00019962108978474263, + "loss": 1.1487, + "step": 36 + }, + { + "epoch": 0.15578947368421053, + "grad_norm": 0.310546875, + "learning_rate": 0.0001995914025736386, + "loss": 1.3543, + "step": 37 + }, + { + "epoch": 0.16, + "grad_norm": 0.306640625, + "learning_rate": 0.00019956059820218982, + "loss": 1.2986, + "step": 38 + }, + { + "epoch": 0.16421052631578947, + "grad_norm": 0.625, + "learning_rate": 0.00019952867701594247, + "loss": 1.0167, + "step": 39 + }, + { + "epoch": 0.16842105263157894, + "grad_norm": 0.306640625, + "learning_rate": 0.00019949563937297045, + "loss": 1.2287, + "step": 40 + }, + { + "epoch": 0.1726315789473684, + "grad_norm": 0.34765625, + "learning_rate": 0.00019946148564387145, + "loss": 1.0063, + "step": 41 + }, + { + "epoch": 0.17684210526315788, + "grad_norm": 0.94921875, + "learning_rate": 0.00019942621621176282, + "loss": 1.2712, + "step": 42 + }, + { + "epoch": 0.18105263157894738, + "grad_norm": 0.333984375, + "learning_rate": 0.00019938983147227716, + "loss": 1.2106, + "step": 43 + }, + { + "epoch": 0.18526315789473685, + "grad_norm": 0.6328125, + "learning_rate": 0.0001993523318335581, + "loss": 1.4887, + "step": 44 + }, + { + "epoch": 0.18947368421052632, + "grad_norm": 0.62890625, + "learning_rate": 0.00019931371771625544, + "loss": 1.2747, + "step": 45 + }, + { + "epoch": 0.1936842105263158, + "grad_norm": 0.2216796875, + "learning_rate": 0.00019927398955352061, + "loss": 1.1946, + "step": 46 + }, + { + "epoch": 0.19789473684210526, + "grad_norm": 0.54296875, + "learning_rate": 0.00019923314779100184, + "loss": 1.2467, + "step": 47 + }, + { + "epoch": 0.20210526315789473, + "grad_norm": 0.25390625, + "learning_rate": 0.00019919119288683908, + "loss": 1.5067, + "step": 48 + }, + { + "epoch": 0.2063157894736842, + "grad_norm": 0.404296875, + "learning_rate": 0.00019914812531165877, + "loss": 1.4291, + "step": 49 + }, + { + "epoch": 0.21052631578947367, + "grad_norm": 0.267578125, + "learning_rate": 0.00019910394554856876, + "loss": 1.0136, + "step": 50 + }, + { + "epoch": 0.21473684210526317, + "grad_norm": 1.328125, + "learning_rate": 0.00019905865409315288, + "loss": 1.2874, + "step": 51 + }, + { + "epoch": 0.21894736842105264, + "grad_norm": 0.244140625, + "learning_rate": 0.0001990122514534651, + "loss": 0.9588, + "step": 52 + }, + { + "epoch": 0.2231578947368421, + "grad_norm": 0.376953125, + "learning_rate": 0.0001989647381500242, + "loss": 1.3036, + "step": 53 + }, + { + "epoch": 0.22736842105263158, + "grad_norm": 0.5, + "learning_rate": 0.00019891611471580764, + "loss": 1.514, + "step": 54 + }, + { + "epoch": 0.23157894736842105, + "grad_norm": 0.423828125, + "learning_rate": 0.0001988663816962458, + "loss": 1.2444, + "step": 55 + }, + { + "epoch": 0.23578947368421052, + "grad_norm": 0.30078125, + "learning_rate": 0.00019881553964921572, + "loss": 1.2261, + "step": 56 + }, + { + "epoch": 0.24, + "grad_norm": 0.34765625, + "learning_rate": 0.00019876358914503492, + "loss": 1.2489, + "step": 57 + }, + { + "epoch": 0.24421052631578946, + "grad_norm": 0.431640625, + "learning_rate": 0.00019871053076645488, + "loss": 1.2134, + "step": 58 + }, + { + "epoch": 0.24842105263157896, + "grad_norm": 0.259765625, + "learning_rate": 0.00019865636510865467, + "loss": 1.1999, + "step": 59 + }, + { + "epoch": 0.25263157894736843, + "grad_norm": 0.2421875, + "learning_rate": 0.00019860109277923418, + "loss": 1.1485, + "step": 60 + }, + { + "epoch": 0.25263157894736843, + "eval_loss": 1.2386109828948975, + "eval_runtime": 25.8708, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.933, + "step": 60 + }, + { + "epoch": 0.25684210526315787, + "grad_norm": 0.5390625, + "learning_rate": 0.00019854471439820733, + "loss": 1.2963, + "step": 61 + }, + { + "epoch": 0.26105263157894737, + "grad_norm": 0.345703125, + "learning_rate": 0.00019848723059799506, + "loss": 1.097, + "step": 62 + }, + { + "epoch": 0.26526315789473687, + "grad_norm": 0.49609375, + "learning_rate": 0.00019842864202341835, + "loss": 1.1463, + "step": 63 + }, + { + "epoch": 0.2694736842105263, + "grad_norm": 0.267578125, + "learning_rate": 0.00019836894933169088, + "loss": 1.1921, + "step": 64 + }, + { + "epoch": 0.2736842105263158, + "grad_norm": 0.349609375, + "learning_rate": 0.00019830815319241168, + "loss": 1.1835, + "step": 65 + }, + { + "epoch": 0.27789473684210525, + "grad_norm": 0.333984375, + "learning_rate": 0.0001982462542875576, + "loss": 1.1998, + "step": 66 + }, + { + "epoch": 0.28210526315789475, + "grad_norm": 0.244140625, + "learning_rate": 0.00019818325331147585, + "loss": 1.1047, + "step": 67 + }, + { + "epoch": 0.2863157894736842, + "grad_norm": 0.244140625, + "learning_rate": 0.00019811915097087587, + "loss": 1.2452, + "step": 68 + }, + { + "epoch": 0.2905263157894737, + "grad_norm": 0.259765625, + "learning_rate": 0.00019805394798482167, + "loss": 1.1633, + "step": 69 + }, + { + "epoch": 0.29473684210526313, + "grad_norm": 0.44140625, + "learning_rate": 0.00019798764508472373, + "loss": 1.0081, + "step": 70 + }, + { + "epoch": 0.29894736842105263, + "grad_norm": 0.5703125, + "learning_rate": 0.00019792024301433064, + "loss": 0.9194, + "step": 71 + }, + { + "epoch": 0.3031578947368421, + "grad_norm": 0.357421875, + "learning_rate": 0.00019785174252972092, + "loss": 0.8091, + "step": 72 + }, + { + "epoch": 0.30736842105263157, + "grad_norm": 0.33984375, + "learning_rate": 0.00019778214439929452, + "loss": 1.1248, + "step": 73 + }, + { + "epoch": 0.31157894736842107, + "grad_norm": 0.23046875, + "learning_rate": 0.0001977114494037641, + "loss": 1.1204, + "step": 74 + }, + { + "epoch": 0.3157894736842105, + "grad_norm": 0.51171875, + "learning_rate": 0.0001976396583361464, + "loss": 1.1662, + "step": 75 + }, + { + "epoch": 0.32, + "grad_norm": 0.2578125, + "learning_rate": 0.00019756677200175315, + "loss": 1.1222, + "step": 76 + }, + { + "epoch": 0.32421052631578945, + "grad_norm": 0.796875, + "learning_rate": 0.00019749279121818235, + "loss": 1.0329, + "step": 77 + }, + { + "epoch": 0.32842105263157895, + "grad_norm": 0.2490234375, + "learning_rate": 0.0001974177168153088, + "loss": 1.1199, + "step": 78 + }, + { + "epoch": 0.33263157894736844, + "grad_norm": 0.26171875, + "learning_rate": 0.00019734154963527495, + "loss": 1.4391, + "step": 79 + }, + { + "epoch": 0.3368421052631579, + "grad_norm": 0.30078125, + "learning_rate": 0.0001972642905324813, + "loss": 1.3924, + "step": 80 + }, + { + "epoch": 0.3410526315789474, + "grad_norm": 0.208984375, + "learning_rate": 0.00019718594037357713, + "loss": 1.1518, + "step": 81 + }, + { + "epoch": 0.3452631578947368, + "grad_norm": 0.953125, + "learning_rate": 0.0001971065000374504, + "loss": 1.2606, + "step": 82 + }, + { + "epoch": 0.3494736842105263, + "grad_norm": 0.306640625, + "learning_rate": 0.0001970259704152181, + "loss": 1.2293, + "step": 83 + }, + { + "epoch": 0.35368421052631577, + "grad_norm": 0.6328125, + "learning_rate": 0.0001969443524102163, + "loss": 0.9564, + "step": 84 + }, + { + "epoch": 0.35789473684210527, + "grad_norm": 0.279296875, + "learning_rate": 0.00019686164693798986, + "loss": 1.2023, + "step": 85 + }, + { + "epoch": 0.36210526315789476, + "grad_norm": 0.263671875, + "learning_rate": 0.0001967778549262822, + "loss": 0.9322, + "step": 86 + }, + { + "epoch": 0.3663157894736842, + "grad_norm": 0.51171875, + "learning_rate": 0.00019669297731502507, + "loss": 1.2589, + "step": 87 + }, + { + "epoch": 0.3705263157894737, + "grad_norm": 0.392578125, + "learning_rate": 0.00019660701505632772, + "loss": 1.3188, + "step": 88 + }, + { + "epoch": 0.37473684210526315, + "grad_norm": 0.51953125, + "learning_rate": 0.00019651996911446647, + "loss": 1.0415, + "step": 89 + }, + { + "epoch": 0.37894736842105264, + "grad_norm": 0.427734375, + "learning_rate": 0.0001964318404658737, + "loss": 1.467, + "step": 90 + }, + { + "epoch": 0.3831578947368421, + "grad_norm": 0.4453125, + "learning_rate": 0.00019634263009912705, + "loss": 1.5527, + "step": 91 + }, + { + "epoch": 0.3873684210526316, + "grad_norm": 0.30859375, + "learning_rate": 0.00019625233901493822, + "loss": 1.2043, + "step": 92 + }, + { + "epoch": 0.391578947368421, + "grad_norm": 0.279296875, + "learning_rate": 0.00019616096822614186, + "loss": 0.9677, + "step": 93 + }, + { + "epoch": 0.3957894736842105, + "grad_norm": 0.46484375, + "learning_rate": 0.000196068518757684, + "loss": 1.1942, + "step": 94 + }, + { + "epoch": 0.4, + "grad_norm": 0.435546875, + "learning_rate": 0.00019597499164661093, + "loss": 1.3481, + "step": 95 + }, + { + "epoch": 0.40421052631578946, + "grad_norm": 0.291015625, + "learning_rate": 0.00019588038794205703, + "loss": 1.2776, + "step": 96 + }, + { + "epoch": 0.40842105263157896, + "grad_norm": 0.3828125, + "learning_rate": 0.00019578470870523359, + "loss": 1.4808, + "step": 97 + }, + { + "epoch": 0.4126315789473684, + "grad_norm": 0.28125, + "learning_rate": 0.00019568795500941635, + "loss": 1.3385, + "step": 98 + }, + { + "epoch": 0.4168421052631579, + "grad_norm": 0.484375, + "learning_rate": 0.00019559012793993392, + "loss": 1.5787, + "step": 99 + }, + { + "epoch": 0.42105263157894735, + "grad_norm": 0.298828125, + "learning_rate": 0.00019549122859415538, + "loss": 1.2151, + "step": 100 + }, + { + "epoch": 0.42526315789473684, + "grad_norm": 0.65234375, + "learning_rate": 0.0001953912580814779, + "loss": 1.0332, + "step": 101 + }, + { + "epoch": 0.42947368421052634, + "grad_norm": 0.37890625, + "learning_rate": 0.00019529021752331453, + "loss": 1.2677, + "step": 102 + }, + { + "epoch": 0.4336842105263158, + "grad_norm": 0.265625, + "learning_rate": 0.0001951881080530814, + "loss": 1.0672, + "step": 103 + }, + { + "epoch": 0.4378947368421053, + "grad_norm": 0.498046875, + "learning_rate": 0.00019508493081618513, + "loss": 1.1736, + "step": 104 + }, + { + "epoch": 0.4421052631578947, + "grad_norm": 0.298828125, + "learning_rate": 0.00019498068697000994, + "loss": 1.1174, + "step": 105 + }, + { + "epoch": 0.4463157894736842, + "grad_norm": 0.169921875, + "learning_rate": 0.00019487537768390464, + "loss": 1.0712, + "step": 106 + }, + { + "epoch": 0.45052631578947366, + "grad_norm": 0.55078125, + "learning_rate": 0.0001947690041391696, + "loss": 1.4496, + "step": 107 + }, + { + "epoch": 0.45473684210526316, + "grad_norm": 0.47265625, + "learning_rate": 0.00019466156752904343, + "loss": 1.2413, + "step": 108 + }, + { + "epoch": 0.4589473684210526, + "grad_norm": 0.6015625, + "learning_rate": 0.0001945530690586896, + "loss": 1.0208, + "step": 109 + }, + { + "epoch": 0.4631578947368421, + "grad_norm": 0.35546875, + "learning_rate": 0.0001944435099451829, + "loss": 1.3997, + "step": 110 + }, + { + "epoch": 0.4673684210526316, + "grad_norm": 0.330078125, + "learning_rate": 0.00019433289141749586, + "loss": 1.2914, + "step": 111 + }, + { + "epoch": 0.47157894736842104, + "grad_norm": 0.63671875, + "learning_rate": 0.00019422121471648497, + "loss": 0.8134, + "step": 112 + }, + { + "epoch": 0.47578947368421054, + "grad_norm": 0.30859375, + "learning_rate": 0.00019410848109487658, + "loss": 1.1282, + "step": 113 + }, + { + "epoch": 0.48, + "grad_norm": 0.5625, + "learning_rate": 0.0001939946918172531, + "loss": 1.3254, + "step": 114 + }, + { + "epoch": 0.4842105263157895, + "grad_norm": 0.30859375, + "learning_rate": 0.00019387984816003867, + "loss": 1.6428, + "step": 115 + }, + { + "epoch": 0.4884210526315789, + "grad_norm": 0.33203125, + "learning_rate": 0.00019376395141148476, + "loss": 1.0688, + "step": 116 + }, + { + "epoch": 0.4926315789473684, + "grad_norm": 0.6953125, + "learning_rate": 0.00019364700287165604, + "loss": 1.0969, + "step": 117 + }, + { + "epoch": 0.4968421052631579, + "grad_norm": 1.296875, + "learning_rate": 0.00019352900385241536, + "loss": 1.5249, + "step": 118 + }, + { + "epoch": 0.5010526315789474, + "grad_norm": 0.2001953125, + "learning_rate": 0.00019340995567740938, + "loss": 1.2095, + "step": 119 + }, + { + "epoch": 0.5052631578947369, + "grad_norm": 0.53515625, + "learning_rate": 0.0001932898596820536, + "loss": 1.1249, + "step": 120 + }, + { + "epoch": 0.5052631578947369, + "eval_loss": 1.2166697978973389, + "eval_runtime": 25.8713, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.933, + "step": 120 + }, + { + "epoch": 0.5094736842105263, + "grad_norm": 0.423828125, + "learning_rate": 0.00019316871721351733, + "loss": 1.5017, + "step": 121 + }, + { + "epoch": 0.5136842105263157, + "grad_norm": 0.7421875, + "learning_rate": 0.0001930465296307087, + "loss": 1.0994, + "step": 122 + }, + { + "epoch": 0.5178947368421053, + "grad_norm": 0.21875, + "learning_rate": 0.0001929232983042592, + "loss": 1.0339, + "step": 123 + }, + { + "epoch": 0.5221052631578947, + "grad_norm": 0.6328125, + "learning_rate": 0.00019279902461650866, + "loss": 1.1317, + "step": 124 + }, + { + "epoch": 0.5263157894736842, + "grad_norm": 0.3984375, + "learning_rate": 0.00019267370996148934, + "loss": 0.9421, + "step": 125 + }, + { + "epoch": 0.5305263157894737, + "grad_norm": 0.396484375, + "learning_rate": 0.00019254735574491058, + "loss": 1.3029, + "step": 126 + }, + { + "epoch": 0.5347368421052632, + "grad_norm": 0.8359375, + "learning_rate": 0.00019241996338414296, + "loss": 1.3548, + "step": 127 + }, + { + "epoch": 0.5389473684210526, + "grad_norm": 0.84765625, + "learning_rate": 0.00019229153430820232, + "loss": 1.0865, + "step": 128 + }, + { + "epoch": 0.5431578947368421, + "grad_norm": 0.453125, + "learning_rate": 0.00019216206995773373, + "loss": 0.8151, + "step": 129 + }, + { + "epoch": 0.5473684210526316, + "grad_norm": 0.8046875, + "learning_rate": 0.0001920315717849956, + "loss": 1.3576, + "step": 130 + }, + { + "epoch": 0.5515789473684211, + "grad_norm": 0.82421875, + "learning_rate": 0.00019190004125384293, + "loss": 1.2439, + "step": 131 + }, + { + "epoch": 0.5557894736842105, + "grad_norm": 0.56640625, + "learning_rate": 0.0001917674798397113, + "loss": 0.9743, + "step": 132 + }, + { + "epoch": 0.56, + "grad_norm": 1.1015625, + "learning_rate": 0.0001916338890296001, + "loss": 1.3315, + "step": 133 + }, + { + "epoch": 0.5642105263157895, + "grad_norm": 0.333984375, + "learning_rate": 0.00019149927032205587, + "loss": 1.3268, + "step": 134 + }, + { + "epoch": 0.5684210526315789, + "grad_norm": 0.388671875, + "learning_rate": 0.00019136362522715562, + "loss": 1.4751, + "step": 135 + }, + { + "epoch": 0.5726315789473684, + "grad_norm": 0.2734375, + "learning_rate": 0.00019122695526648968, + "loss": 1.1219, + "step": 136 + }, + { + "epoch": 0.5768421052631579, + "grad_norm": 1.0703125, + "learning_rate": 0.00019108926197314486, + "loss": 1.4059, + "step": 137 + }, + { + "epoch": 0.5810526315789474, + "grad_norm": 0.33203125, + "learning_rate": 0.00019095054689168705, + "loss": 1.3282, + "step": 138 + }, + { + "epoch": 0.5852631578947368, + "grad_norm": 0.6171875, + "learning_rate": 0.00019081081157814404, + "loss": 1.6334, + "step": 139 + }, + { + "epoch": 0.5894736842105263, + "grad_norm": 0.359375, + "learning_rate": 0.00019067005759998797, + "loss": 1.159, + "step": 140 + }, + { + "epoch": 0.5936842105263158, + "grad_norm": 0.470703125, + "learning_rate": 0.00019052828653611786, + "loss": 1.3261, + "step": 141 + }, + { + "epoch": 0.5978947368421053, + "grad_norm": 0.26171875, + "learning_rate": 0.0001903854999768417, + "loss": 1.2041, + "step": 142 + }, + { + "epoch": 0.6021052631578947, + "grad_norm": 0.27734375, + "learning_rate": 0.00019024169952385885, + "loss": 1.1998, + "step": 143 + }, + { + "epoch": 0.6063157894736843, + "grad_norm": 0.56640625, + "learning_rate": 0.0001900968867902419, + "loss": 1.6427, + "step": 144 + }, + { + "epoch": 0.6105263157894737, + "grad_norm": 0.419921875, + "learning_rate": 0.00018995106340041873, + "loss": 1.1968, + "step": 145 + }, + { + "epoch": 0.6147368421052631, + "grad_norm": 0.671875, + "learning_rate": 0.00018980423099015402, + "loss": 0.8641, + "step": 146 + }, + { + "epoch": 0.6189473684210526, + "grad_norm": 0.333984375, + "learning_rate": 0.00018965639120653121, + "loss": 1.4518, + "step": 147 + }, + { + "epoch": 0.6231578947368421, + "grad_norm": 0.423828125, + "learning_rate": 0.00018950754570793384, + "loss": 1.3077, + "step": 148 + }, + { + "epoch": 0.6273684210526316, + "grad_norm": 0.85546875, + "learning_rate": 0.00018935769616402703, + "loss": 1.2077, + "step": 149 + }, + { + "epoch": 0.631578947368421, + "grad_norm": 0.224609375, + "learning_rate": 0.00018920684425573865, + "loss": 1.0868, + "step": 150 + }, + { + "epoch": 0.6357894736842106, + "grad_norm": 0.25390625, + "learning_rate": 0.00018905499167524062, + "loss": 1.2018, + "step": 151 + }, + { + "epoch": 0.64, + "grad_norm": 0.498046875, + "learning_rate": 0.00018890214012592975, + "loss": 1.1675, + "step": 152 + }, + { + "epoch": 0.6442105263157895, + "grad_norm": 0.2451171875, + "learning_rate": 0.00018874829132240874, + "loss": 1.2348, + "step": 153 + }, + { + "epoch": 0.6484210526315789, + "grad_norm": 0.33984375, + "learning_rate": 0.000188593446990467, + "loss": 1.3202, + "step": 154 + }, + { + "epoch": 0.6526315789473685, + "grad_norm": 1.0078125, + "learning_rate": 0.0001884376088670611, + "loss": 1.0171, + "step": 155 + }, + { + "epoch": 0.6568421052631579, + "grad_norm": 0.318359375, + "learning_rate": 0.00018828077870029552, + "loss": 1.219, + "step": 156 + }, + { + "epoch": 0.6610526315789473, + "grad_norm": 0.369140625, + "learning_rate": 0.00018812295824940285, + "loss": 0.8956, + "step": 157 + }, + { + "epoch": 0.6652631578947369, + "grad_norm": 0.328125, + "learning_rate": 0.00018796414928472417, + "loss": 0.9322, + "step": 158 + }, + { + "epoch": 0.6694736842105263, + "grad_norm": 0.2890625, + "learning_rate": 0.0001878043535876891, + "loss": 1.2062, + "step": 159 + }, + { + "epoch": 0.6736842105263158, + "grad_norm": 0.41015625, + "learning_rate": 0.0001876435729507959, + "loss": 1.3113, + "step": 160 + }, + { + "epoch": 0.6778947368421052, + "grad_norm": 0.53515625, + "learning_rate": 0.00018748180917759136, + "loss": 1.1539, + "step": 161 + }, + { + "epoch": 0.6821052631578948, + "grad_norm": 0.279296875, + "learning_rate": 0.0001873190640826505, + "loss": 1.3287, + "step": 162 + }, + { + "epoch": 0.6863157894736842, + "grad_norm": 0.298828125, + "learning_rate": 0.00018715533949155625, + "loss": 1.0616, + "step": 163 + }, + { + "epoch": 0.6905263157894737, + "grad_norm": 0.388671875, + "learning_rate": 0.00018699063724087904, + "loss": 1.0171, + "step": 164 + }, + { + "epoch": 0.6947368421052632, + "grad_norm": 0.32421875, + "learning_rate": 0.00018682495917815608, + "loss": 1.1113, + "step": 165 + }, + { + "epoch": 0.6989473684210527, + "grad_norm": 0.46875, + "learning_rate": 0.00018665830716187065, + "loss": 0.8351, + "step": 166 + }, + { + "epoch": 0.7031578947368421, + "grad_norm": 0.6015625, + "learning_rate": 0.0001864906830614314, + "loss": 1.2181, + "step": 167 + }, + { + "epoch": 0.7073684210526315, + "grad_norm": 0.71875, + "learning_rate": 0.0001863220887571512, + "loss": 1.1843, + "step": 168 + }, + { + "epoch": 0.7115789473684211, + "grad_norm": 0.51953125, + "learning_rate": 0.00018615252614022617, + "loss": 1.3809, + "step": 169 + }, + { + "epoch": 0.7157894736842105, + "grad_norm": 0.34375, + "learning_rate": 0.0001859819971127143, + "loss": 1.4288, + "step": 170 + }, + { + "epoch": 0.72, + "grad_norm": 0.41796875, + "learning_rate": 0.00018581050358751445, + "loss": 1.1449, + "step": 171 + }, + { + "epoch": 0.7242105263157895, + "grad_norm": 0.578125, + "learning_rate": 0.00018563804748834438, + "loss": 1.0703, + "step": 172 + }, + { + "epoch": 0.728421052631579, + "grad_norm": 0.43359375, + "learning_rate": 0.00018546463074971977, + "loss": 1.2093, + "step": 173 + }, + { + "epoch": 0.7326315789473684, + "grad_norm": 0.26171875, + "learning_rate": 0.000185290255316932, + "loss": 0.7346, + "step": 174 + }, + { + "epoch": 0.7368421052631579, + "grad_norm": 0.369140625, + "learning_rate": 0.00018511492314602656, + "loss": 1.2782, + "step": 175 + }, + { + "epoch": 0.7410526315789474, + "grad_norm": 0.56640625, + "learning_rate": 0.00018493863620378122, + "loss": 1.0844, + "step": 176 + }, + { + "epoch": 0.7452631578947368, + "grad_norm": 0.2470703125, + "learning_rate": 0.00018476139646768372, + "loss": 1.1771, + "step": 177 + }, + { + "epoch": 0.7494736842105263, + "grad_norm": 0.228515625, + "learning_rate": 0.00018458320592590975, + "loss": 1.2, + "step": 178 + }, + { + "epoch": 0.7536842105263157, + "grad_norm": 0.310546875, + "learning_rate": 0.00018440406657730058, + "loss": 1.2248, + "step": 179 + }, + { + "epoch": 0.7578947368421053, + "grad_norm": 0.263671875, + "learning_rate": 0.00018422398043134067, + "loss": 1.3675, + "step": 180 + }, + { + "epoch": 0.7578947368421053, + "eval_loss": 1.2129631042480469, + "eval_runtime": 25.9253, + "eval_samples_per_second": 3.857, + "eval_steps_per_second": 1.929, + "step": 180 + }, + { + "epoch": 0.7621052631578947, + "grad_norm": 0.2412109375, + "learning_rate": 0.0001840429495081352, + "loss": 1.2456, + "step": 181 + }, + { + "epoch": 0.7663157894736842, + "grad_norm": 0.25390625, + "learning_rate": 0.00018386097583838714, + "loss": 1.2617, + "step": 182 + }, + { + "epoch": 0.7705263157894737, + "grad_norm": 0.283203125, + "learning_rate": 0.00018367806146337483, + "loss": 1.2708, + "step": 183 + }, + { + "epoch": 0.7747368421052632, + "grad_norm": 0.318359375, + "learning_rate": 0.00018349420843492888, + "loss": 1.3771, + "step": 184 + }, + { + "epoch": 0.7789473684210526, + "grad_norm": 0.73046875, + "learning_rate": 0.00018330941881540915, + "loss": 1.3123, + "step": 185 + }, + { + "epoch": 0.783157894736842, + "grad_norm": 0.46875, + "learning_rate": 0.00018312369467768166, + "loss": 1.1952, + "step": 186 + }, + { + "epoch": 0.7873684210526316, + "grad_norm": 0.7578125, + "learning_rate": 0.00018293703810509537, + "loss": 1.426, + "step": 187 + }, + { + "epoch": 0.791578947368421, + "grad_norm": 0.33203125, + "learning_rate": 0.0001827494511914587, + "loss": 1.3205, + "step": 188 + }, + { + "epoch": 0.7957894736842105, + "grad_norm": 0.365234375, + "learning_rate": 0.00018256093604101617, + "loss": 1.2221, + "step": 189 + }, + { + "epoch": 0.8, + "grad_norm": 0.36328125, + "learning_rate": 0.0001823714947684247, + "loss": 1.4212, + "step": 190 + }, + { + "epoch": 0.8042105263157895, + "grad_norm": 0.349609375, + "learning_rate": 0.00018218112949872995, + "loss": 1.2183, + "step": 191 + }, + { + "epoch": 0.8084210526315789, + "grad_norm": 0.392578125, + "learning_rate": 0.00018198984236734246, + "loss": 0.9922, + "step": 192 + }, + { + "epoch": 0.8126315789473684, + "grad_norm": 0.373046875, + "learning_rate": 0.0001817976355200137, + "loss": 1.334, + "step": 193 + }, + { + "epoch": 0.8168421052631579, + "grad_norm": 0.30859375, + "learning_rate": 0.000181604511112812, + "loss": 1.1052, + "step": 194 + }, + { + "epoch": 0.8210526315789474, + "grad_norm": 0.349609375, + "learning_rate": 0.0001814104713120983, + "loss": 0.8112, + "step": 195 + }, + { + "epoch": 0.8252631578947368, + "grad_norm": 0.35546875, + "learning_rate": 0.000181215518294502, + "loss": 1.2507, + "step": 196 + }, + { + "epoch": 0.8294736842105264, + "grad_norm": 0.671875, + "learning_rate": 0.0001810196542468964, + "loss": 1.1228, + "step": 197 + }, + { + "epoch": 0.8336842105263158, + "grad_norm": 0.41015625, + "learning_rate": 0.00018082288136637422, + "loss": 1.3562, + "step": 198 + }, + { + "epoch": 0.8378947368421052, + "grad_norm": 0.302734375, + "learning_rate": 0.000180625201860223, + "loss": 0.7766, + "step": 199 + }, + { + "epoch": 0.8421052631578947, + "grad_norm": 0.322265625, + "learning_rate": 0.00018042661794590023, + "loss": 1.2885, + "step": 200 + }, + { + "epoch": 0.8463157894736842, + "grad_norm": 0.609375, + "learning_rate": 0.0001802271318510086, + "loss": 1.271, + "step": 201 + }, + { + "epoch": 0.8505263157894737, + "grad_norm": 0.341796875, + "learning_rate": 0.00018002674581327094, + "loss": 1.3777, + "step": 202 + }, + { + "epoch": 0.8547368421052631, + "grad_norm": 0.3828125, + "learning_rate": 0.00017982546208050515, + "loss": 1.32, + "step": 203 + }, + { + "epoch": 0.8589473684210527, + "grad_norm": 0.392578125, + "learning_rate": 0.00017962328291059888, + "loss": 1.1265, + "step": 204 + }, + { + "epoch": 0.8631578947368421, + "grad_norm": 0.73046875, + "learning_rate": 0.00017942021057148436, + "loss": 1.523, + "step": 205 + }, + { + "epoch": 0.8673684210526316, + "grad_norm": 1.046875, + "learning_rate": 0.00017921624734111292, + "loss": 1.0251, + "step": 206 + }, + { + "epoch": 0.871578947368421, + "grad_norm": 0.490234375, + "learning_rate": 0.0001790113955074294, + "loss": 1.2665, + "step": 207 + }, + { + "epoch": 0.8757894736842106, + "grad_norm": 0.6875, + "learning_rate": 0.0001788056573683464, + "loss": 1.2244, + "step": 208 + }, + { + "epoch": 0.88, + "grad_norm": 0.2177734375, + "learning_rate": 0.0001785990352317187, + "loss": 1.1621, + "step": 209 + }, + { + "epoch": 0.8842105263157894, + "grad_norm": 0.4921875, + "learning_rate": 0.00017839153141531718, + "loss": 1.3182, + "step": 210 + }, + { + "epoch": 0.888421052631579, + "grad_norm": 0.388671875, + "learning_rate": 0.000178183148246803, + "loss": 1.4508, + "step": 211 + }, + { + "epoch": 0.8926315789473684, + "grad_norm": 0.388671875, + "learning_rate": 0.00017797388806370132, + "loss": 1.1155, + "step": 212 + }, + { + "epoch": 0.8968421052631579, + "grad_norm": 0.416015625, + "learning_rate": 0.00017776375321337521, + "loss": 0.9191, + "step": 213 + }, + { + "epoch": 0.9010526315789473, + "grad_norm": 0.2431640625, + "learning_rate": 0.00017755274605299923, + "loss": 1.1577, + "step": 214 + }, + { + "epoch": 0.9052631578947369, + "grad_norm": 0.412109375, + "learning_rate": 0.00017734086894953305, + "loss": 1.372, + "step": 215 + }, + { + "epoch": 0.9094736842105263, + "grad_norm": 0.353515625, + "learning_rate": 0.00017712812427969485, + "loss": 1.1932, + "step": 216 + }, + { + "epoch": 0.9136842105263158, + "grad_norm": 0.291015625, + "learning_rate": 0.0001769145144299347, + "loss": 1.3167, + "step": 217 + }, + { + "epoch": 0.9178947368421052, + "grad_norm": 0.255859375, + "learning_rate": 0.00017670004179640774, + "loss": 1.2357, + "step": 218 + }, + { + "epoch": 0.9221052631578948, + "grad_norm": 0.5703125, + "learning_rate": 0.00017648470878494737, + "loss": 0.9556, + "step": 219 + }, + { + "epoch": 0.9263157894736842, + "grad_norm": 0.36328125, + "learning_rate": 0.0001762685178110382, + "loss": 1.2047, + "step": 220 + }, + { + "epoch": 0.9305263157894736, + "grad_norm": 0.271484375, + "learning_rate": 0.00017605147129978903, + "loss": 1.1935, + "step": 221 + }, + { + "epoch": 0.9347368421052632, + "grad_norm": 0.400390625, + "learning_rate": 0.0001758335716859055, + "loss": 1.3426, + "step": 222 + }, + { + "epoch": 0.9389473684210526, + "grad_norm": 0.73046875, + "learning_rate": 0.000175614821413663, + "loss": 1.5181, + "step": 223 + }, + { + "epoch": 0.9431578947368421, + "grad_norm": 0.3203125, + "learning_rate": 0.00017539522293687898, + "loss": 1.0962, + "step": 224 + }, + { + "epoch": 0.9473684210526315, + "grad_norm": 0.232421875, + "learning_rate": 0.00017517477871888568, + "loss": 1.3922, + "step": 225 + }, + { + "epoch": 0.9515789473684211, + "grad_norm": 0.5546875, + "learning_rate": 0.00017495349123250242, + "loss": 1.2535, + "step": 226 + }, + { + "epoch": 0.9557894736842105, + "grad_norm": 0.296875, + "learning_rate": 0.00017473136296000772, + "loss": 1.1987, + "step": 227 + }, + { + "epoch": 0.96, + "grad_norm": 0.77734375, + "learning_rate": 0.00017450839639311162, + "loss": 1.4301, + "step": 228 + }, + { + "epoch": 0.9642105263157895, + "grad_norm": 0.35546875, + "learning_rate": 0.00017428459403292767, + "loss": 1.3164, + "step": 229 + }, + { + "epoch": 0.968421052631579, + "grad_norm": 0.419921875, + "learning_rate": 0.00017405995838994494, + "loss": 1.1554, + "step": 230 + }, + { + "epoch": 0.9726315789473684, + "grad_norm": 0.3125, + "learning_rate": 0.00017383449198399964, + "loss": 1.3036, + "step": 231 + }, + { + "epoch": 0.9768421052631578, + "grad_norm": 0.267578125, + "learning_rate": 0.00017360819734424715, + "loss": 1.2797, + "step": 232 + }, + { + "epoch": 0.9810526315789474, + "grad_norm": 0.412109375, + "learning_rate": 0.00017338107700913354, + "loss": 1.1202, + "step": 233 + }, + { + "epoch": 0.9852631578947368, + "grad_norm": 0.365234375, + "learning_rate": 0.0001731531335263669, + "loss": 0.8187, + "step": 234 + }, + { + "epoch": 0.9894736842105263, + "grad_norm": 0.5546875, + "learning_rate": 0.0001729243694528891, + "loss": 1.1702, + "step": 235 + }, + { + "epoch": 0.9936842105263158, + "grad_norm": 0.5546875, + "learning_rate": 0.00017269478735484683, + "loss": 1.1406, + "step": 236 + }, + { + "epoch": 0.9978947368421053, + "grad_norm": 0.3359375, + "learning_rate": 0.00017246438980756293, + "loss": 0.9633, + "step": 237 + }, + { + "epoch": 1.0021052631578948, + "grad_norm": 0.359375, + "learning_rate": 0.00017223317939550753, + "loss": 1.2181, + "step": 238 + }, + { + "epoch": 1.0063157894736843, + "grad_norm": 0.2138671875, + "learning_rate": 0.00017200115871226904, + "loss": 1.2035, + "step": 239 + }, + { + "epoch": 1.0105263157894737, + "grad_norm": 0.3203125, + "learning_rate": 0.00017176833036052495, + "loss": 1.3449, + "step": 240 + }, + { + "epoch": 1.0105263157894737, + "eval_loss": 1.1994891166687012, + "eval_runtime": 25.9787, + "eval_samples_per_second": 3.849, + "eval_steps_per_second": 1.925, + "step": 240 + }, + { + "epoch": 1.0147368421052632, + "grad_norm": 0.375, + "learning_rate": 0.00017153469695201277, + "loss": 1.0695, + "step": 241 + }, + { + "epoch": 1.0189473684210526, + "grad_norm": 0.578125, + "learning_rate": 0.0001713002611075007, + "loss": 0.8898, + "step": 242 + }, + { + "epoch": 1.023157894736842, + "grad_norm": 0.203125, + "learning_rate": 0.0001710650254567582, + "loss": 1.087, + "step": 243 + }, + { + "epoch": 1.0273684210526315, + "grad_norm": 0.3828125, + "learning_rate": 0.0001708289926385265, + "loss": 1.1739, + "step": 244 + }, + { + "epoch": 1.0315789473684212, + "grad_norm": 0.3515625, + "learning_rate": 0.000170592165300489, + "loss": 1.0413, + "step": 245 + }, + { + "epoch": 1.0357894736842106, + "grad_norm": 0.376953125, + "learning_rate": 0.0001703545460992416, + "loss": 1.0745, + "step": 246 + }, + { + "epoch": 1.04, + "grad_norm": 0.25, + "learning_rate": 0.00017011613770026292, + "loss": 1.2494, + "step": 247 + }, + { + "epoch": 1.0442105263157895, + "grad_norm": 0.419921875, + "learning_rate": 0.00016987694277788417, + "loss": 0.7568, + "step": 248 + }, + { + "epoch": 1.048421052631579, + "grad_norm": 0.6015625, + "learning_rate": 0.0001696369640152596, + "loss": 1.2675, + "step": 249 + }, + { + "epoch": 1.0526315789473684, + "grad_norm": 0.67578125, + "learning_rate": 0.0001693962041043359, + "loss": 1.0456, + "step": 250 + }, + { + "epoch": 1.0568421052631578, + "grad_norm": 0.5234375, + "learning_rate": 0.00016915466574582246, + "loss": 1.1145, + "step": 251 + }, + { + "epoch": 1.0610526315789475, + "grad_norm": 0.279296875, + "learning_rate": 0.00016891235164916065, + "loss": 1.1659, + "step": 252 + }, + { + "epoch": 1.065263157894737, + "grad_norm": 0.369140625, + "learning_rate": 0.0001686692645324938, + "loss": 0.9993, + "step": 253 + }, + { + "epoch": 1.0694736842105264, + "grad_norm": 0.78125, + "learning_rate": 0.00016842540712263637, + "loss": 1.2847, + "step": 254 + }, + { + "epoch": 1.0736842105263158, + "grad_norm": 0.48046875, + "learning_rate": 0.0001681807821550438, + "loss": 1.1483, + "step": 255 + }, + { + "epoch": 1.0778947368421052, + "grad_norm": 0.73046875, + "learning_rate": 0.00016793539237378128, + "loss": 0.7895, + "step": 256 + }, + { + "epoch": 1.0821052631578947, + "grad_norm": 0.30859375, + "learning_rate": 0.00016768924053149338, + "loss": 1.2321, + "step": 257 + }, + { + "epoch": 1.0863157894736841, + "grad_norm": 0.291015625, + "learning_rate": 0.00016744232938937308, + "loss": 1.0098, + "step": 258 + }, + { + "epoch": 1.0905263157894738, + "grad_norm": 0.470703125, + "learning_rate": 0.00016719466171713073, + "loss": 1.3362, + "step": 259 + }, + { + "epoch": 1.0947368421052632, + "grad_norm": 0.2890625, + "learning_rate": 0.0001669462402929629, + "loss": 1.2487, + "step": 260 + }, + { + "epoch": 1.0989473684210527, + "grad_norm": 0.39453125, + "learning_rate": 0.00016669706790352148, + "loss": 1.1732, + "step": 261 + }, + { + "epoch": 1.1031578947368421, + "grad_norm": 0.28125, + "learning_rate": 0.00016644714734388217, + "loss": 1.1493, + "step": 262 + }, + { + "epoch": 1.1073684210526316, + "grad_norm": 0.314453125, + "learning_rate": 0.0001661964814175133, + "loss": 1.0521, + "step": 263 + }, + { + "epoch": 1.111578947368421, + "grad_norm": 0.53125, + "learning_rate": 0.00016594507293624425, + "loss": 1.07, + "step": 264 + }, + { + "epoch": 1.1157894736842104, + "grad_norm": 0.2890625, + "learning_rate": 0.00016569292472023392, + "loss": 1.1393, + "step": 265 + }, + { + "epoch": 1.12, + "grad_norm": 0.390625, + "learning_rate": 0.00016544003959793925, + "loss": 1.2059, + "step": 266 + }, + { + "epoch": 1.1242105263157895, + "grad_norm": 0.494140625, + "learning_rate": 0.0001651864204060833, + "loss": 0.9832, + "step": 267 + }, + { + "epoch": 1.128421052631579, + "grad_norm": 0.376953125, + "learning_rate": 0.00016493206998962354, + "loss": 0.9154, + "step": 268 + }, + { + "epoch": 1.1326315789473684, + "grad_norm": 0.39453125, + "learning_rate": 0.00016467699120171987, + "loss": 1.0839, + "step": 269 + }, + { + "epoch": 1.1368421052631579, + "grad_norm": 0.94140625, + "learning_rate": 0.0001644211869037027, + "loss": 1.1298, + "step": 270 + }, + { + "epoch": 1.1410526315789473, + "grad_norm": 0.341796875, + "learning_rate": 0.00016416465996504077, + "loss": 1.131, + "step": 271 + }, + { + "epoch": 1.1452631578947368, + "grad_norm": 0.90234375, + "learning_rate": 0.00016390741326330907, + "loss": 1.0148, + "step": 272 + }, + { + "epoch": 1.1494736842105264, + "grad_norm": 0.35546875, + "learning_rate": 0.00016364944968415632, + "loss": 0.788, + "step": 273 + }, + { + "epoch": 1.1536842105263159, + "grad_norm": 0.294921875, + "learning_rate": 0.00016339077212127294, + "loss": 0.9585, + "step": 274 + }, + { + "epoch": 1.1578947368421053, + "grad_norm": 0.56640625, + "learning_rate": 0.00016313138347635837, + "loss": 0.9937, + "step": 275 + }, + { + "epoch": 1.1621052631578948, + "grad_norm": 0.251953125, + "learning_rate": 0.0001628712866590885, + "loss": 1.069, + "step": 276 + }, + { + "epoch": 1.1663157894736842, + "grad_norm": 0.384765625, + "learning_rate": 0.00016261048458708318, + "loss": 1.2734, + "step": 277 + }, + { + "epoch": 1.1705263157894736, + "grad_norm": 0.66015625, + "learning_rate": 0.00016234898018587337, + "loss": 1.1955, + "step": 278 + }, + { + "epoch": 1.174736842105263, + "grad_norm": 0.3984375, + "learning_rate": 0.00016208677638886835, + "loss": 0.9949, + "step": 279 + }, + { + "epoch": 1.1789473684210527, + "grad_norm": 0.33984375, + "learning_rate": 0.00016182387613732291, + "loss": 1.1192, + "step": 280 + }, + { + "epoch": 1.1831578947368422, + "grad_norm": 0.4765625, + "learning_rate": 0.00016156028238030422, + "loss": 1.165, + "step": 281 + }, + { + "epoch": 1.1873684210526316, + "grad_norm": 0.51171875, + "learning_rate": 0.00016129599807465875, + "loss": 0.9109, + "step": 282 + }, + { + "epoch": 1.191578947368421, + "grad_norm": 0.5546875, + "learning_rate": 0.00016103102618497922, + "loss": 0.9441, + "step": 283 + }, + { + "epoch": 1.1957894736842105, + "grad_norm": 0.7109375, + "learning_rate": 0.0001607653696835713, + "loss": 1.3679, + "step": 284 + }, + { + "epoch": 1.2, + "grad_norm": 0.3671875, + "learning_rate": 0.0001604990315504202, + "loss": 0.8203, + "step": 285 + }, + { + "epoch": 1.2042105263157894, + "grad_norm": 0.51171875, + "learning_rate": 0.00016023201477315731, + "loss": 1.052, + "step": 286 + }, + { + "epoch": 1.208421052631579, + "grad_norm": 0.72265625, + "learning_rate": 0.00015996432234702666, + "loss": 1.281, + "step": 287 + }, + { + "epoch": 1.2126315789473685, + "grad_norm": 0.44921875, + "learning_rate": 0.0001596959572748514, + "loss": 1.1851, + "step": 288 + }, + { + "epoch": 1.216842105263158, + "grad_norm": 1.3515625, + "learning_rate": 0.0001594269225669999, + "loss": 1.1378, + "step": 289 + }, + { + "epoch": 1.2210526315789474, + "grad_norm": 0.337890625, + "learning_rate": 0.00015915722124135227, + "loss": 0.7495, + "step": 290 + }, + { + "epoch": 1.2252631578947368, + "grad_norm": 0.45703125, + "learning_rate": 0.00015888685632326627, + "loss": 1.0089, + "step": 291 + }, + { + "epoch": 1.2294736842105263, + "grad_norm": 0.470703125, + "learning_rate": 0.00015861583084554349, + "loss": 0.7145, + "step": 292 + }, + { + "epoch": 1.2336842105263157, + "grad_norm": 0.4375, + "learning_rate": 0.00015834414784839532, + "loss": 1.1728, + "step": 293 + }, + { + "epoch": 1.2378947368421054, + "grad_norm": 1.5703125, + "learning_rate": 0.0001580718103794089, + "loss": 1.351, + "step": 294 + }, + { + "epoch": 1.2421052631578948, + "grad_norm": 0.6171875, + "learning_rate": 0.0001577988214935127, + "loss": 1.0116, + "step": 295 + }, + { + "epoch": 1.2463157894736843, + "grad_norm": 0.392578125, + "learning_rate": 0.00015752518425294257, + "loss": 1.1591, + "step": 296 + }, + { + "epoch": 1.2505263157894737, + "grad_norm": 0.44140625, + "learning_rate": 0.0001572509017272072, + "loss": 1.0848, + "step": 297 + }, + { + "epoch": 1.2547368421052632, + "grad_norm": 0.43359375, + "learning_rate": 0.00015697597699305366, + "loss": 1.0024, + "step": 298 + }, + { + "epoch": 1.2589473684210526, + "grad_norm": 0.470703125, + "learning_rate": 0.00015670041313443304, + "loss": 1.072, + "step": 299 + }, + { + "epoch": 1.263157894736842, + "grad_norm": 0.34375, + "learning_rate": 0.00015642421324246568, + "loss": 1.1825, + "step": 300 + }, + { + "epoch": 1.263157894736842, + "eval_loss": 1.207429051399231, + "eval_runtime": 25.9189, + "eval_samples_per_second": 3.858, + "eval_steps_per_second": 1.929, + "step": 300 + }, + { + "epoch": 1.2673684210526317, + "grad_norm": 1.0234375, + "learning_rate": 0.00015614738041540668, + "loss": 0.8984, + "step": 301 + }, + { + "epoch": 1.271578947368421, + "grad_norm": 0.29296875, + "learning_rate": 0.00015586991775861102, + "loss": 0.9569, + "step": 302 + }, + { + "epoch": 1.2757894736842106, + "grad_norm": 0.59375, + "learning_rate": 0.00015559182838449873, + "loss": 1.177, + "step": 303 + }, + { + "epoch": 1.28, + "grad_norm": 0.416015625, + "learning_rate": 0.00015531311541251995, + "loss": 0.9957, + "step": 304 + }, + { + "epoch": 1.2842105263157895, + "grad_norm": 0.47265625, + "learning_rate": 0.00015503378196912013, + "loss": 0.8083, + "step": 305 + }, + { + "epoch": 1.288421052631579, + "grad_norm": 0.7734375, + "learning_rate": 0.00015475383118770472, + "loss": 1.0463, + "step": 306 + }, + { + "epoch": 1.2926315789473684, + "grad_norm": 0.30859375, + "learning_rate": 0.00015447326620860415, + "loss": 1.0709, + "step": 307 + }, + { + "epoch": 1.296842105263158, + "grad_norm": 0.35546875, + "learning_rate": 0.00015419209017903852, + "loss": 1.2547, + "step": 308 + }, + { + "epoch": 1.3010526315789472, + "grad_norm": 0.3671875, + "learning_rate": 0.00015391030625308249, + "loss": 0.8027, + "step": 309 + }, + { + "epoch": 1.305263157894737, + "grad_norm": 0.28515625, + "learning_rate": 0.0001536279175916296, + "loss": 1.0735, + "step": 310 + }, + { + "epoch": 1.3094736842105263, + "grad_norm": 0.625, + "learning_rate": 0.00015334492736235705, + "loss": 1.0046, + "step": 311 + }, + { + "epoch": 1.3136842105263158, + "grad_norm": 0.83203125, + "learning_rate": 0.0001530613387396901, + "loss": 1.1327, + "step": 312 + }, + { + "epoch": 1.3178947368421052, + "grad_norm": 0.515625, + "learning_rate": 0.00015277715490476641, + "loss": 0.9917, + "step": 313 + }, + { + "epoch": 1.3221052631578947, + "grad_norm": 0.359375, + "learning_rate": 0.0001524923790454004, + "loss": 1.2532, + "step": 314 + }, + { + "epoch": 1.3263157894736843, + "grad_norm": 0.306640625, + "learning_rate": 0.0001522070143560476, + "loss": 0.9906, + "step": 315 + }, + { + "epoch": 1.3305263157894736, + "grad_norm": 0.33984375, + "learning_rate": 0.00015192106403776848, + "loss": 1.0958, + "step": 316 + }, + { + "epoch": 1.3347368421052632, + "grad_norm": 0.33203125, + "learning_rate": 0.000151634531298193, + "loss": 1.1674, + "step": 317 + }, + { + "epoch": 1.3389473684210527, + "grad_norm": 0.45703125, + "learning_rate": 0.0001513474193514842, + "loss": 0.9189, + "step": 318 + }, + { + "epoch": 1.343157894736842, + "grad_norm": 0.55078125, + "learning_rate": 0.0001510597314183025, + "loss": 1.2956, + "step": 319 + }, + { + "epoch": 1.3473684210526315, + "grad_norm": 0.41015625, + "learning_rate": 0.00015077147072576933, + "loss": 1.3053, + "step": 320 + }, + { + "epoch": 1.351578947368421, + "grad_norm": 0.421875, + "learning_rate": 0.00015048264050743096, + "loss": 1.26, + "step": 321 + }, + { + "epoch": 1.3557894736842107, + "grad_norm": 0.318359375, + "learning_rate": 0.00015019324400322243, + "loss": 1.1314, + "step": 322 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 0.33203125, + "learning_rate": 0.00014990328445943092, + "loss": 1.1536, + "step": 323 + }, + { + "epoch": 1.3642105263157895, + "grad_norm": 0.365234375, + "learning_rate": 0.00014961276512865954, + "loss": 0.8678, + "step": 324 + }, + { + "epoch": 1.368421052631579, + "grad_norm": 0.56640625, + "learning_rate": 0.00014932168926979074, + "loss": 1.0508, + "step": 325 + }, + { + "epoch": 1.3726315789473684, + "grad_norm": 0.333984375, + "learning_rate": 0.00014903006014794983, + "loss": 0.9275, + "step": 326 + }, + { + "epoch": 1.3768421052631579, + "grad_norm": 0.61328125, + "learning_rate": 0.00014873788103446825, + "loss": 1.0225, + "step": 327 + }, + { + "epoch": 1.3810526315789473, + "grad_norm": 0.357421875, + "learning_rate": 0.00014844515520684703, + "loss": 0.9955, + "step": 328 + }, + { + "epoch": 1.385263157894737, + "grad_norm": 0.6328125, + "learning_rate": 0.00014815188594871983, + "loss": 1.2857, + "step": 329 + }, + { + "epoch": 1.3894736842105262, + "grad_norm": 0.5078125, + "learning_rate": 0.00014785807654981627, + "loss": 1.0598, + "step": 330 + }, + { + "epoch": 1.3936842105263159, + "grad_norm": 0.96484375, + "learning_rate": 0.00014756373030592497, + "loss": 1.1031, + "step": 331 + }, + { + "epoch": 1.3978947368421053, + "grad_norm": 0.380859375, + "learning_rate": 0.00014726885051885653, + "loss": 1.3287, + "step": 332 + }, + { + "epoch": 1.4021052631578947, + "grad_norm": 0.5859375, + "learning_rate": 0.00014697344049640658, + "loss": 1.1952, + "step": 333 + }, + { + "epoch": 1.4063157894736842, + "grad_norm": 0.453125, + "learning_rate": 0.0001466775035523186, + "loss": 1.083, + "step": 334 + }, + { + "epoch": 1.4105263157894736, + "grad_norm": 0.37890625, + "learning_rate": 0.00014638104300624685, + "loss": 0.8393, + "step": 335 + }, + { + "epoch": 1.4147368421052633, + "grad_norm": 0.9921875, + "learning_rate": 0.00014608406218371894, + "loss": 1.3947, + "step": 336 + }, + { + "epoch": 1.4189473684210525, + "grad_norm": 0.357421875, + "learning_rate": 0.0001457865644160987, + "loss": 1.028, + "step": 337 + }, + { + "epoch": 1.4231578947368422, + "grad_norm": 0.671875, + "learning_rate": 0.00014548855304054886, + "loss": 1.1156, + "step": 338 + }, + { + "epoch": 1.4273684210526316, + "grad_norm": 0.6953125, + "learning_rate": 0.00014519003139999337, + "loss": 1.1537, + "step": 339 + }, + { + "epoch": 1.431578947368421, + "grad_norm": 0.408203125, + "learning_rate": 0.00014489100284308017, + "loss": 1.207, + "step": 340 + }, + { + "epoch": 1.4357894736842105, + "grad_norm": 0.56640625, + "learning_rate": 0.00014459147072414335, + "loss": 1.125, + "step": 341 + }, + { + "epoch": 1.44, + "grad_norm": 0.423828125, + "learning_rate": 0.00014429143840316585, + "loss": 1.0774, + "step": 342 + }, + { + "epoch": 1.4442105263157896, + "grad_norm": 1.0859375, + "learning_rate": 0.0001439909092457415, + "loss": 1.3349, + "step": 343 + }, + { + "epoch": 1.4484210526315788, + "grad_norm": 0.55078125, + "learning_rate": 0.00014368988662303732, + "loss": 0.6138, + "step": 344 + }, + { + "epoch": 1.4526315789473685, + "grad_norm": 0.330078125, + "learning_rate": 0.00014338837391175582, + "loss": 1.1961, + "step": 345 + }, + { + "epoch": 1.456842105263158, + "grad_norm": 0.54296875, + "learning_rate": 0.00014308637449409706, + "loss": 1.067, + "step": 346 + }, + { + "epoch": 1.4610526315789474, + "grad_norm": 0.294921875, + "learning_rate": 0.0001427838917577206, + "loss": 1.0033, + "step": 347 + }, + { + "epoch": 1.4652631578947368, + "grad_norm": 0.494140625, + "learning_rate": 0.00014248092909570774, + "loss": 1.011, + "step": 348 + }, + { + "epoch": 1.4694736842105263, + "grad_norm": 0.75, + "learning_rate": 0.00014217748990652318, + "loss": 1.3504, + "step": 349 + }, + { + "epoch": 1.4736842105263157, + "grad_norm": 0.41796875, + "learning_rate": 0.00014187357759397714, + "loss": 1.0421, + "step": 350 + }, + { + "epoch": 1.4778947368421052, + "grad_norm": 0.73828125, + "learning_rate": 0.00014156919556718698, + "loss": 1.1829, + "step": 351 + }, + { + "epoch": 1.4821052631578948, + "grad_norm": 0.31640625, + "learning_rate": 0.00014126434724053913, + "loss": 1.1985, + "step": 352 + }, + { + "epoch": 1.4863157894736843, + "grad_norm": 0.41015625, + "learning_rate": 0.00014095903603365066, + "loss": 1.1609, + "step": 353 + }, + { + "epoch": 1.4905263157894737, + "grad_norm": 0.345703125, + "learning_rate": 0.00014065326537133094, + "loss": 1.2552, + "step": 354 + }, + { + "epoch": 1.4947368421052631, + "grad_norm": 1.2265625, + "learning_rate": 0.0001403470386835434, + "loss": 0.9655, + "step": 355 + }, + { + "epoch": 1.4989473684210526, + "grad_norm": 0.46875, + "learning_rate": 0.0001400403594053667, + "loss": 0.9709, + "step": 356 + }, + { + "epoch": 1.5031578947368422, + "grad_norm": 0.703125, + "learning_rate": 0.00013973323097695657, + "loss": 1.048, + "step": 357 + }, + { + "epoch": 1.5073684210526315, + "grad_norm": 0.494140625, + "learning_rate": 0.00013942565684350698, + "loss": 1.0853, + "step": 358 + }, + { + "epoch": 1.5115789473684211, + "grad_norm": 0.9140625, + "learning_rate": 0.00013911764045521159, + "loss": 1.1191, + "step": 359 + }, + { + "epoch": 1.5157894736842106, + "grad_norm": 0.478515625, + "learning_rate": 0.00013880918526722497, + "loss": 0.9782, + "step": 360 + }, + { + "epoch": 1.5157894736842106, + "eval_loss": 1.2060081958770752, + "eval_runtime": 25.9425, + "eval_samples_per_second": 3.855, + "eval_steps_per_second": 1.927, + "step": 360 + }, + { + "epoch": 1.52, + "grad_norm": 0.361328125, + "learning_rate": 0.00013850029473962402, + "loss": 0.9822, + "step": 361 + }, + { + "epoch": 1.5242105263157895, + "grad_norm": 0.44921875, + "learning_rate": 0.00013819097233736888, + "loss": 0.957, + "step": 362 + }, + { + "epoch": 1.528421052631579, + "grad_norm": 0.478515625, + "learning_rate": 0.0001378812215302644, + "loss": 0.9774, + "step": 363 + }, + { + "epoch": 1.5326315789473686, + "grad_norm": 0.35546875, + "learning_rate": 0.00013757104579292082, + "loss": 1.2854, + "step": 364 + }, + { + "epoch": 1.5368421052631578, + "grad_norm": 0.671875, + "learning_rate": 0.00013726044860471522, + "loss": 0.9209, + "step": 365 + }, + { + "epoch": 1.5410526315789475, + "grad_norm": 0.400390625, + "learning_rate": 0.00013694943344975212, + "loss": 0.9489, + "step": 366 + }, + { + "epoch": 1.545263157894737, + "grad_norm": 0.625, + "learning_rate": 0.00013663800381682464, + "loss": 1.2956, + "step": 367 + }, + { + "epoch": 1.5494736842105263, + "grad_norm": 0.3203125, + "learning_rate": 0.00013632616319937522, + "loss": 1.0642, + "step": 368 + }, + { + "epoch": 1.5536842105263158, + "grad_norm": 0.296875, + "learning_rate": 0.0001360139150954566, + "loss": 1.034, + "step": 369 + }, + { + "epoch": 1.5578947368421052, + "grad_norm": 0.5234375, + "learning_rate": 0.00013570126300769232, + "loss": 0.8314, + "step": 370 + }, + { + "epoch": 1.5621052631578949, + "grad_norm": 0.279296875, + "learning_rate": 0.00013538821044323772, + "loss": 1.1934, + "step": 371 + }, + { + "epoch": 1.566315789473684, + "grad_norm": 0.353515625, + "learning_rate": 0.0001350747609137404, + "loss": 1.2661, + "step": 372 + }, + { + "epoch": 1.5705263157894738, + "grad_norm": 0.27734375, + "learning_rate": 0.00013476091793530092, + "loss": 0.9746, + "step": 373 + }, + { + "epoch": 1.5747368421052632, + "grad_norm": 0.49609375, + "learning_rate": 0.0001344466850284333, + "loss": 1.0562, + "step": 374 + }, + { + "epoch": 1.5789473684210527, + "grad_norm": 0.27734375, + "learning_rate": 0.00013413206571802558, + "loss": 0.8612, + "step": 375 + }, + { + "epoch": 1.583157894736842, + "grad_norm": 0.51953125, + "learning_rate": 0.00013381706353330014, + "loss": 0.7912, + "step": 376 + }, + { + "epoch": 1.5873684210526315, + "grad_norm": 0.45703125, + "learning_rate": 0.00013350168200777448, + "loss": 1.0244, + "step": 377 + }, + { + "epoch": 1.5915789473684212, + "grad_norm": 0.5703125, + "learning_rate": 0.0001331859246792211, + "loss": 1.1267, + "step": 378 + }, + { + "epoch": 1.5957894736842104, + "grad_norm": 0.48828125, + "learning_rate": 0.00013286979508962815, + "loss": 1.0055, + "step": 379 + }, + { + "epoch": 1.6, + "grad_norm": 0.310546875, + "learning_rate": 0.0001325532967851596, + "loss": 1.2348, + "step": 380 + }, + { + "epoch": 1.6042105263157893, + "grad_norm": 0.369140625, + "learning_rate": 0.00013223643331611537, + "loss": 0.991, + "step": 381 + }, + { + "epoch": 1.608421052631579, + "grad_norm": 0.51171875, + "learning_rate": 0.00013191920823689177, + "loss": 1.0261, + "step": 382 + }, + { + "epoch": 1.6126315789473684, + "grad_norm": 0.90234375, + "learning_rate": 0.00013160162510594123, + "loss": 1.0752, + "step": 383 + }, + { + "epoch": 1.6168421052631579, + "grad_norm": 1.1796875, + "learning_rate": 0.00013128368748573273, + "loss": 1.1808, + "step": 384 + }, + { + "epoch": 1.6210526315789475, + "grad_norm": 0.40625, + "learning_rate": 0.0001309653989427117, + "loss": 0.8969, + "step": 385 + }, + { + "epoch": 1.6252631578947367, + "grad_norm": 0.53515625, + "learning_rate": 0.00013064676304726, + "loss": 1.1139, + "step": 386 + }, + { + "epoch": 1.6294736842105264, + "grad_norm": 0.478515625, + "learning_rate": 0.00013032778337365592, + "loss": 1.2781, + "step": 387 + }, + { + "epoch": 1.6336842105263156, + "grad_norm": 0.66796875, + "learning_rate": 0.0001300084635000341, + "loss": 1.1434, + "step": 388 + }, + { + "epoch": 1.6378947368421053, + "grad_norm": 0.33984375, + "learning_rate": 0.00012968880700834517, + "loss": 0.7428, + "step": 389 + }, + { + "epoch": 1.6421052631578947, + "grad_norm": 0.357421875, + "learning_rate": 0.000129368817484316, + "loss": 1.256, + "step": 390 + }, + { + "epoch": 1.6463157894736842, + "grad_norm": 0.384765625, + "learning_rate": 0.00012904849851740906, + "loss": 1.1516, + "step": 391 + }, + { + "epoch": 1.6505263157894738, + "grad_norm": 0.62890625, + "learning_rate": 0.0001287278537007824, + "loss": 1.1683, + "step": 392 + }, + { + "epoch": 1.654736842105263, + "grad_norm": 0.4140625, + "learning_rate": 0.0001284068866312492, + "loss": 0.9682, + "step": 393 + }, + { + "epoch": 1.6589473684210527, + "grad_norm": 0.36328125, + "learning_rate": 0.00012808560090923758, + "loss": 1.2109, + "step": 394 + }, + { + "epoch": 1.663157894736842, + "grad_norm": 0.3125, + "learning_rate": 0.00012776400013875006, + "loss": 1.1139, + "step": 395 + }, + { + "epoch": 1.6673684210526316, + "grad_norm": 0.453125, + "learning_rate": 0.00012744208792732324, + "loss": 1.0977, + "step": 396 + }, + { + "epoch": 1.671578947368421, + "grad_norm": 0.44140625, + "learning_rate": 0.00012711986788598725, + "loss": 1.1872, + "step": 397 + }, + { + "epoch": 1.6757894736842105, + "grad_norm": 0.59765625, + "learning_rate": 0.00012679734362922528, + "loss": 1.0773, + "step": 398 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 0.265625, + "learning_rate": 0.00012647451877493307, + "loss": 0.6203, + "step": 399 + }, + { + "epoch": 1.6842105263157894, + "grad_norm": 0.5078125, + "learning_rate": 0.00012615139694437835, + "loss": 1.1557, + "step": 400 + }, + { + "epoch": 1.688421052631579, + "grad_norm": 0.578125, + "learning_rate": 0.00012582798176215997, + "loss": 1.1022, + "step": 401 + }, + { + "epoch": 1.6926315789473683, + "grad_norm": 0.4609375, + "learning_rate": 0.00012550427685616765, + "loss": 1.193, + "step": 402 + }, + { + "epoch": 1.696842105263158, + "grad_norm": 0.455078125, + "learning_rate": 0.00012518028585754094, + "loss": 0.8758, + "step": 403 + }, + { + "epoch": 1.7010526315789474, + "grad_norm": 0.51171875, + "learning_rate": 0.00012485601240062869, + "loss": 0.9792, + "step": 404 + }, + { + "epoch": 1.7052631578947368, + "grad_norm": 0.3046875, + "learning_rate": 0.00012453146012294812, + "loss": 1.1347, + "step": 405 + }, + { + "epoch": 1.7094736842105265, + "grad_norm": 0.439453125, + "learning_rate": 0.00012420663266514417, + "loss": 1.1692, + "step": 406 + }, + { + "epoch": 1.7136842105263157, + "grad_norm": 0.33203125, + "learning_rate": 0.00012388153367094858, + "loss": 0.9904, + "step": 407 + }, + { + "epoch": 1.7178947368421054, + "grad_norm": 0.62109375, + "learning_rate": 0.0001235561667871391, + "loss": 0.8854, + "step": 408 + }, + { + "epoch": 1.7221052631578946, + "grad_norm": 0.52734375, + "learning_rate": 0.00012323053566349834, + "loss": 1.3143, + "step": 409 + }, + { + "epoch": 1.7263157894736842, + "grad_norm": 0.390625, + "learning_rate": 0.0001229046439527732, + "loss": 1.0957, + "step": 410 + }, + { + "epoch": 1.7305263157894737, + "grad_norm": 0.494140625, + "learning_rate": 0.00012257849531063352, + "loss": 1.0458, + "step": 411 + }, + { + "epoch": 1.7347368421052631, + "grad_norm": 0.34375, + "learning_rate": 0.00012225209339563145, + "loss": 1.3262, + "step": 412 + }, + { + "epoch": 1.7389473684210528, + "grad_norm": 0.32421875, + "learning_rate": 0.00012192544186916006, + "loss": 1.1506, + "step": 413 + }, + { + "epoch": 1.743157894736842, + "grad_norm": 0.671875, + "learning_rate": 0.00012159854439541245, + "loss": 1.035, + "step": 414 + }, + { + "epoch": 1.7473684210526317, + "grad_norm": 0.4453125, + "learning_rate": 0.00012127140464134066, + "loss": 1.3485, + "step": 415 + }, + { + "epoch": 1.751578947368421, + "grad_norm": 0.439453125, + "learning_rate": 0.00012094402627661447, + "loss": 1.1842, + "step": 416 + }, + { + "epoch": 1.7557894736842106, + "grad_norm": 0.51171875, + "learning_rate": 0.00012061641297358022, + "loss": 1.4594, + "step": 417 + }, + { + "epoch": 1.76, + "grad_norm": 0.34375, + "learning_rate": 0.00012028856840721974, + "loss": 1.1014, + "step": 418 + }, + { + "epoch": 1.7642105263157895, + "grad_norm": 0.640625, + "learning_rate": 0.00011996049625510901, + "loss": 0.7333, + "step": 419 + }, + { + "epoch": 1.768421052631579, + "grad_norm": 0.5859375, + "learning_rate": 0.00011963220019737691, + "loss": 1.2063, + "step": 420 + }, + { + "epoch": 1.768421052631579, + "eval_loss": 1.199410080909729, + "eval_runtime": 25.8809, + "eval_samples_per_second": 3.864, + "eval_steps_per_second": 1.932, + "step": 420 + }, + { + "epoch": 1.7726315789473683, + "grad_norm": 0.73046875, + "learning_rate": 0.00011930368391666398, + "loss": 1.2112, + "step": 421 + }, + { + "epoch": 1.776842105263158, + "grad_norm": 0.67578125, + "learning_rate": 0.00011897495109808107, + "loss": 1.0342, + "step": 422 + }, + { + "epoch": 1.7810526315789472, + "grad_norm": 0.376953125, + "learning_rate": 0.00011864600542916813, + "loss": 1.2239, + "step": 423 + }, + { + "epoch": 1.7852631578947369, + "grad_norm": 0.361328125, + "learning_rate": 0.00011831685059985262, + "loss": 0.9817, + "step": 424 + }, + { + "epoch": 1.7894736842105263, + "grad_norm": 0.30859375, + "learning_rate": 0.00011798749030240834, + "loss": 1.3617, + "step": 425 + }, + { + "epoch": 1.7936842105263158, + "grad_norm": 0.7265625, + "learning_rate": 0.00011765792823141384, + "loss": 1.2746, + "step": 426 + }, + { + "epoch": 1.7978947368421052, + "grad_norm": 0.484375, + "learning_rate": 0.00011732816808371115, + "loss": 1.471, + "step": 427 + }, + { + "epoch": 1.8021052631578947, + "grad_norm": 0.349609375, + "learning_rate": 0.00011699821355836409, + "loss": 1.1071, + "step": 428 + }, + { + "epoch": 1.8063157894736843, + "grad_norm": 0.26953125, + "learning_rate": 0.00011666806835661699, + "loss": 1.0702, + "step": 429 + }, + { + "epoch": 1.8105263157894735, + "grad_norm": 0.31640625, + "learning_rate": 0.00011633773618185302, + "loss": 1.2166, + "step": 430 + }, + { + "epoch": 1.8147368421052632, + "grad_norm": 0.36328125, + "learning_rate": 0.00011600722073955282, + "loss": 1.1926, + "step": 431 + }, + { + "epoch": 1.8189473684210526, + "grad_norm": 0.35546875, + "learning_rate": 0.00011567652573725262, + "loss": 1.0472, + "step": 432 + }, + { + "epoch": 1.823157894736842, + "grad_norm": 0.765625, + "learning_rate": 0.00011534565488450311, + "loss": 0.9258, + "step": 433 + }, + { + "epoch": 1.8273684210526315, + "grad_norm": 0.765625, + "learning_rate": 0.00011501461189282733, + "loss": 1.6986, + "step": 434 + }, + { + "epoch": 1.831578947368421, + "grad_norm": 0.392578125, + "learning_rate": 0.00011468340047567947, + "loss": 1.1726, + "step": 435 + }, + { + "epoch": 1.8357894736842106, + "grad_norm": 0.412109375, + "learning_rate": 0.00011435202434840287, + "loss": 1.2373, + "step": 436 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 0.6015625, + "learning_rate": 0.0001140204872281886, + "loss": 0.7262, + "step": 437 + }, + { + "epoch": 1.8442105263157895, + "grad_norm": 0.490234375, + "learning_rate": 0.0001136887928340336, + "loss": 1.0527, + "step": 438 + }, + { + "epoch": 1.848421052631579, + "grad_norm": 0.453125, + "learning_rate": 0.00011335694488669904, + "loss": 0.89, + "step": 439 + }, + { + "epoch": 1.8526315789473684, + "grad_norm": 0.55859375, + "learning_rate": 0.00011302494710866857, + "loss": 1.109, + "step": 440 + }, + { + "epoch": 1.8568421052631578, + "grad_norm": 0.5546875, + "learning_rate": 0.00011269280322410658, + "loss": 1.0818, + "step": 441 + }, + { + "epoch": 1.8610526315789473, + "grad_norm": 0.3671875, + "learning_rate": 0.00011236051695881633, + "loss": 1.2366, + "step": 442 + }, + { + "epoch": 1.865263157894737, + "grad_norm": 0.427734375, + "learning_rate": 0.0001120280920401983, + "loss": 1.1401, + "step": 443 + }, + { + "epoch": 1.8694736842105262, + "grad_norm": 0.5, + "learning_rate": 0.00011169553219720828, + "loss": 1.0061, + "step": 444 + }, + { + "epoch": 1.8736842105263158, + "grad_norm": 0.43359375, + "learning_rate": 0.0001113628411603155, + "loss": 1.2742, + "step": 445 + }, + { + "epoch": 1.8778947368421053, + "grad_norm": 0.375, + "learning_rate": 0.00011103002266146096, + "loss": 1.296, + "step": 446 + }, + { + "epoch": 1.8821052631578947, + "grad_norm": 0.7578125, + "learning_rate": 0.0001106970804340154, + "loss": 0.9062, + "step": 447 + }, + { + "epoch": 1.8863157894736842, + "grad_norm": 0.44140625, + "learning_rate": 0.0001103640182127375, + "loss": 0.85, + "step": 448 + }, + { + "epoch": 1.8905263157894736, + "grad_norm": 0.6171875, + "learning_rate": 0.00011003083973373187, + "loss": 1.0045, + "step": 449 + }, + { + "epoch": 1.8947368421052633, + "grad_norm": 0.50390625, + "learning_rate": 0.00010969754873440743, + "loss": 0.8616, + "step": 450 + }, + { + "epoch": 1.8989473684210525, + "grad_norm": 0.5546875, + "learning_rate": 0.0001093641489534351, + "loss": 0.9867, + "step": 451 + }, + { + "epoch": 1.9031578947368422, + "grad_norm": 0.462890625, + "learning_rate": 0.00010903064413070612, + "loss": 1.1836, + "step": 452 + }, + { + "epoch": 1.9073684210526316, + "grad_norm": 0.515625, + "learning_rate": 0.00010869703800729, + "loss": 0.9996, + "step": 453 + }, + { + "epoch": 1.911578947368421, + "grad_norm": 0.5078125, + "learning_rate": 0.00010836333432539272, + "loss": 1.4083, + "step": 454 + }, + { + "epoch": 1.9157894736842105, + "grad_norm": 0.474609375, + "learning_rate": 0.0001080295368283144, + "loss": 0.9889, + "step": 455 + }, + { + "epoch": 1.92, + "grad_norm": 0.30078125, + "learning_rate": 0.00010769564926040769, + "loss": 0.8322, + "step": 456 + }, + { + "epoch": 1.9242105263157896, + "grad_norm": 1.171875, + "learning_rate": 0.00010736167536703556, + "loss": 0.8793, + "step": 457 + }, + { + "epoch": 1.9284210526315788, + "grad_norm": 0.59765625, + "learning_rate": 0.0001070276188945293, + "loss": 1.1383, + "step": 458 + }, + { + "epoch": 1.9326315789473685, + "grad_norm": 0.625, + "learning_rate": 0.00010669348359014657, + "loss": 1.2608, + "step": 459 + }, + { + "epoch": 1.936842105263158, + "grad_norm": 0.49609375, + "learning_rate": 0.00010635927320202928, + "loss": 1.1444, + "step": 460 + }, + { + "epoch": 1.9410526315789474, + "grad_norm": 0.6796875, + "learning_rate": 0.00010602499147916159, + "loss": 1.1779, + "step": 461 + }, + { + "epoch": 1.9452631578947368, + "grad_norm": 0.390625, + "learning_rate": 0.00010569064217132791, + "loss": 1.2066, + "step": 462 + }, + { + "epoch": 1.9494736842105262, + "grad_norm": 0.32421875, + "learning_rate": 0.00010535622902907072, + "loss": 1.3065, + "step": 463 + }, + { + "epoch": 1.953684210526316, + "grad_norm": 0.45703125, + "learning_rate": 0.00010502175580364857, + "loss": 0.9167, + "step": 464 + }, + { + "epoch": 1.9578947368421051, + "grad_norm": 0.8203125, + "learning_rate": 0.00010468722624699401, + "loss": 1.059, + "step": 465 + }, + { + "epoch": 1.9621052631578948, + "grad_norm": 0.93359375, + "learning_rate": 0.00010435264411167148, + "loss": 0.8437, + "step": 466 + }, + { + "epoch": 1.9663157894736842, + "grad_norm": 0.4453125, + "learning_rate": 0.00010401801315083519, + "loss": 1.1105, + "step": 467 + }, + { + "epoch": 1.9705263157894737, + "grad_norm": 0.81640625, + "learning_rate": 0.0001036833371181871, + "loss": 1.5699, + "step": 468 + }, + { + "epoch": 1.9747368421052631, + "grad_norm": 0.359375, + "learning_rate": 0.00010334861976793465, + "loss": 1.0942, + "step": 469 + }, + { + "epoch": 1.9789473684210526, + "grad_norm": 0.4765625, + "learning_rate": 0.00010301386485474889, + "loss": 0.923, + "step": 470 + }, + { + "epoch": 1.9831578947368422, + "grad_norm": 0.498046875, + "learning_rate": 0.00010267907613372217, + "loss": 1.123, + "step": 471 + }, + { + "epoch": 1.9873684210526315, + "grad_norm": 0.45703125, + "learning_rate": 0.00010234425736032607, + "loss": 1.1807, + "step": 472 + }, + { + "epoch": 1.9915789473684211, + "grad_norm": 0.53515625, + "learning_rate": 0.00010200941229036926, + "loss": 1.2465, + "step": 473 + }, + { + "epoch": 1.9957894736842106, + "grad_norm": 0.3984375, + "learning_rate": 0.00010167454467995549, + "loss": 0.9394, + "step": 474 + }, + { + "epoch": 2.0, + "grad_norm": 1.0625, + "learning_rate": 0.00010133965828544128, + "loss": 1.0323, + "step": 475 + }, + { + "epoch": 2.0042105263157897, + "grad_norm": 0.318359375, + "learning_rate": 0.00010100475686339379, + "loss": 1.1268, + "step": 476 + }, + { + "epoch": 2.008421052631579, + "grad_norm": 0.416015625, + "learning_rate": 0.00010066984417054887, + "loss": 0.7652, + "step": 477 + }, + { + "epoch": 2.0126315789473685, + "grad_norm": 0.41015625, + "learning_rate": 0.00010033492396376878, + "loss": 1.0744, + "step": 478 + }, + { + "epoch": 2.0168421052631578, + "grad_norm": 0.296875, + "learning_rate": 0.0001, + "loss": 1.0926, + "step": 479 + }, + { + "epoch": 2.0210526315789474, + "grad_norm": 0.3671875, + "learning_rate": 9.966507603623125e-05, + "loss": 0.9614, + "step": 480 + }, + { + "epoch": 2.0210526315789474, + "eval_loss": 1.1929166316986084, + "eval_runtime": 25.8683, + "eval_samples_per_second": 3.866, + "eval_steps_per_second": 1.933, + "step": 480 + }, + { + "epoch": 2.0252631578947367, + "grad_norm": 0.48046875, + "learning_rate": 9.933015582945112e-05, + "loss": 1.0848, + "step": 481 + }, + { + "epoch": 2.0294736842105263, + "grad_norm": 0.361328125, + "learning_rate": 9.899524313660624e-05, + "loss": 1.0067, + "step": 482 + }, + { + "epoch": 2.033684210526316, + "grad_norm": 0.4375, + "learning_rate": 9.866034171455876e-05, + "loss": 0.7833, + "step": 483 + }, + { + "epoch": 2.037894736842105, + "grad_norm": 0.33984375, + "learning_rate": 9.832545532004454e-05, + "loss": 0.9732, + "step": 484 + }, + { + "epoch": 2.042105263157895, + "grad_norm": 0.341796875, + "learning_rate": 9.799058770963073e-05, + "loss": 0.6746, + "step": 485 + }, + { + "epoch": 2.046315789473684, + "grad_norm": 0.439453125, + "learning_rate": 9.765574263967396e-05, + "loss": 0.7767, + "step": 486 + }, + { + "epoch": 2.0505263157894738, + "grad_norm": 0.54296875, + "learning_rate": 9.732092386627787e-05, + "loss": 0.7498, + "step": 487 + }, + { + "epoch": 2.054736842105263, + "grad_norm": 0.55859375, + "learning_rate": 9.698613514525116e-05, + "loss": 0.6573, + "step": 488 + }, + { + "epoch": 2.0589473684210526, + "grad_norm": 0.40234375, + "learning_rate": 9.665138023206536e-05, + "loss": 0.6588, + "step": 489 + }, + { + "epoch": 2.0631578947368423, + "grad_norm": 0.37109375, + "learning_rate": 9.631666288181293e-05, + "loss": 0.9873, + "step": 490 + }, + { + "epoch": 2.0673684210526315, + "grad_norm": 0.44921875, + "learning_rate": 9.598198684916483e-05, + "loss": 0.7607, + "step": 491 + }, + { + "epoch": 2.071578947368421, + "grad_norm": 0.41015625, + "learning_rate": 9.564735588832856e-05, + "loss": 1.1862, + "step": 492 + }, + { + "epoch": 2.0757894736842104, + "grad_norm": 0.6328125, + "learning_rate": 9.531277375300599e-05, + "loss": 0.9608, + "step": 493 + }, + { + "epoch": 2.08, + "grad_norm": 0.82421875, + "learning_rate": 9.497824419635144e-05, + "loss": 0.6041, + "step": 494 + }, + { + "epoch": 2.0842105263157893, + "grad_norm": 0.60546875, + "learning_rate": 9.464377097092932e-05, + "loss": 1.4274, + "step": 495 + }, + { + "epoch": 2.088421052631579, + "grad_norm": 0.8671875, + "learning_rate": 9.430935782867212e-05, + "loss": 0.6098, + "step": 496 + }, + { + "epoch": 2.0926315789473686, + "grad_norm": 0.58984375, + "learning_rate": 9.397500852083841e-05, + "loss": 1.0759, + "step": 497 + }, + { + "epoch": 2.096842105263158, + "grad_norm": 0.5546875, + "learning_rate": 9.364072679797073e-05, + "loss": 0.8507, + "step": 498 + }, + { + "epoch": 2.1010526315789475, + "grad_norm": 1.09375, + "learning_rate": 9.330651640985346e-05, + "loss": 1.2054, + "step": 499 + }, + { + "epoch": 2.1052631578947367, + "grad_norm": 0.490234375, + "learning_rate": 9.297238110547074e-05, + "loss": 1.0956, + "step": 500 + }, + { + "epoch": 2.1094736842105264, + "grad_norm": 0.482421875, + "learning_rate": 9.263832463296445e-05, + "loss": 1.0093, + "step": 501 + }, + { + "epoch": 2.1136842105263156, + "grad_norm": 0.73046875, + "learning_rate": 9.230435073959232e-05, + "loss": 1.0679, + "step": 502 + }, + { + "epoch": 2.1178947368421053, + "grad_norm": 0.54296875, + "learning_rate": 9.197046317168563e-05, + "loss": 0.8422, + "step": 503 + }, + { + "epoch": 2.122105263157895, + "grad_norm": 0.66015625, + "learning_rate": 9.163666567460733e-05, + "loss": 0.6545, + "step": 504 + }, + { + "epoch": 2.126315789473684, + "grad_norm": 0.578125, + "learning_rate": 9.130296199270999e-05, + "loss": 1.0327, + "step": 505 + }, + { + "epoch": 2.130526315789474, + "grad_norm": 0.392578125, + "learning_rate": 9.096935586929392e-05, + "loss": 0.9649, + "step": 506 + }, + { + "epoch": 2.134736842105263, + "grad_norm": 0.63671875, + "learning_rate": 9.063585104656493e-05, + "loss": 0.6553, + "step": 507 + }, + { + "epoch": 2.1389473684210527, + "grad_norm": 0.4765625, + "learning_rate": 9.030245126559262e-05, + "loss": 0.6973, + "step": 508 + }, + { + "epoch": 2.143157894736842, + "grad_norm": 0.48828125, + "learning_rate": 8.996916026626812e-05, + "loss": 0.7395, + "step": 509 + }, + { + "epoch": 2.1473684210526316, + "grad_norm": 0.70703125, + "learning_rate": 8.963598178726254e-05, + "loss": 1.0481, + "step": 510 + }, + { + "epoch": 2.1515789473684213, + "grad_norm": 0.80859375, + "learning_rate": 8.930291956598462e-05, + "loss": 0.9903, + "step": 511 + }, + { + "epoch": 2.1557894736842105, + "grad_norm": 0.451171875, + "learning_rate": 8.896997733853903e-05, + "loss": 1.2016, + "step": 512 + }, + { + "epoch": 2.16, + "grad_norm": 0.3828125, + "learning_rate": 8.863715883968451e-05, + "loss": 1.2436, + "step": 513 + }, + { + "epoch": 2.1642105263157894, + "grad_norm": 0.333984375, + "learning_rate": 8.830446780279176e-05, + "loss": 0.9862, + "step": 514 + }, + { + "epoch": 2.168421052631579, + "grad_norm": 0.5078125, + "learning_rate": 8.797190795980172e-05, + "loss": 0.5605, + "step": 515 + }, + { + "epoch": 2.1726315789473682, + "grad_norm": 0.6484375, + "learning_rate": 8.763948304118368e-05, + "loss": 0.4958, + "step": 516 + }, + { + "epoch": 2.176842105263158, + "grad_norm": 0.5625, + "learning_rate": 8.730719677589344e-05, + "loss": 1.0046, + "step": 517 + }, + { + "epoch": 2.1810526315789476, + "grad_norm": 0.486328125, + "learning_rate": 8.697505289133145e-05, + "loss": 0.6935, + "step": 518 + }, + { + "epoch": 2.185263157894737, + "grad_norm": 0.37890625, + "learning_rate": 8.6643055113301e-05, + "loss": 1.2277, + "step": 519 + }, + { + "epoch": 2.1894736842105265, + "grad_norm": 0.73828125, + "learning_rate": 8.631120716596641e-05, + "loss": 0.7949, + "step": 520 + }, + { + "epoch": 2.1936842105263157, + "grad_norm": 0.6640625, + "learning_rate": 8.597951277181142e-05, + "loss": 0.8309, + "step": 521 + }, + { + "epoch": 2.1978947368421053, + "grad_norm": 0.58984375, + "learning_rate": 8.564797565159714e-05, + "loss": 0.6418, + "step": 522 + }, + { + "epoch": 2.2021052631578946, + "grad_norm": 0.77734375, + "learning_rate": 8.531659952432056e-05, + "loss": 1.1851, + "step": 523 + }, + { + "epoch": 2.2063157894736842, + "grad_norm": 0.380859375, + "learning_rate": 8.498538810717267e-05, + "loss": 1.015, + "step": 524 + }, + { + "epoch": 2.2105263157894735, + "grad_norm": 0.384765625, + "learning_rate": 8.465434511549691e-05, + "loss": 1.0365, + "step": 525 + }, + { + "epoch": 2.214736842105263, + "grad_norm": 0.41015625, + "learning_rate": 8.432347426274739e-05, + "loss": 1.0889, + "step": 526 + }, + { + "epoch": 2.218947368421053, + "grad_norm": 0.7578125, + "learning_rate": 8.399277926044724e-05, + "loss": 1.0203, + "step": 527 + }, + { + "epoch": 2.223157894736842, + "grad_norm": 0.55078125, + "learning_rate": 8.366226381814697e-05, + "loss": 0.9986, + "step": 528 + }, + { + "epoch": 2.2273684210526317, + "grad_norm": 0.41015625, + "learning_rate": 8.333193164338304e-05, + "loss": 1.1371, + "step": 529 + }, + { + "epoch": 2.231578947368421, + "grad_norm": 0.70703125, + "learning_rate": 8.300178644163594e-05, + "loss": 1.1068, + "step": 530 + }, + { + "epoch": 2.2357894736842105, + "grad_norm": 0.8046875, + "learning_rate": 8.267183191628889e-05, + "loss": 1.0577, + "step": 531 + }, + { + "epoch": 2.24, + "grad_norm": 0.75, + "learning_rate": 8.234207176858614e-05, + "loss": 0.9573, + "step": 532 + }, + { + "epoch": 2.2442105263157894, + "grad_norm": 0.400390625, + "learning_rate": 8.201250969759168e-05, + "loss": 1.4236, + "step": 533 + }, + { + "epoch": 2.248421052631579, + "grad_norm": 0.61328125, + "learning_rate": 8.16831494001474e-05, + "loss": 0.5257, + "step": 534 + }, + { + "epoch": 2.2526315789473683, + "grad_norm": 0.515625, + "learning_rate": 8.13539945708319e-05, + "loss": 1.2213, + "step": 535 + }, + { + "epoch": 2.256842105263158, + "grad_norm": 0.6640625, + "learning_rate": 8.102504890191892e-05, + "loss": 0.8878, + "step": 536 + }, + { + "epoch": 2.261052631578947, + "grad_norm": 0.435546875, + "learning_rate": 8.069631608333605e-05, + "loss": 0.8702, + "step": 537 + }, + { + "epoch": 2.265263157894737, + "grad_norm": 0.482421875, + "learning_rate": 8.036779980262311e-05, + "loss": 0.8551, + "step": 538 + }, + { + "epoch": 2.269473684210526, + "grad_norm": 0.5234375, + "learning_rate": 8.003950374489101e-05, + "loss": 0.6752, + "step": 539 + }, + { + "epoch": 2.2736842105263158, + "grad_norm": 0.546875, + "learning_rate": 7.971143159278026e-05, + "loss": 1.0084, + "step": 540 + }, + { + "epoch": 2.2736842105263158, + "eval_loss": 1.2140343189239502, + "eval_runtime": 25.8753, + "eval_samples_per_second": 3.865, + "eval_steps_per_second": 1.932, + "step": 540 + }, + { + "epoch": 2.2778947368421054, + "grad_norm": 0.349609375, + "learning_rate": 7.938358702641979e-05, + "loss": 1.091, + "step": 541 + }, + { + "epoch": 2.2821052631578946, + "grad_norm": 0.375, + "learning_rate": 7.905597372338558e-05, + "loss": 0.8771, + "step": 542 + }, + { + "epoch": 2.2863157894736843, + "grad_norm": 0.75, + "learning_rate": 7.872859535865935e-05, + "loss": 1.0554, + "step": 543 + }, + { + "epoch": 2.2905263157894735, + "grad_norm": 0.4140625, + "learning_rate": 7.840145560458756e-05, + "loss": 0.958, + "step": 544 + }, + { + "epoch": 2.294736842105263, + "grad_norm": 0.4375, + "learning_rate": 7.807455813083996e-05, + "loss": 0.9956, + "step": 545 + }, + { + "epoch": 2.298947368421053, + "grad_norm": 0.50390625, + "learning_rate": 7.774790660436858e-05, + "loss": 0.864, + "step": 546 + }, + { + "epoch": 2.303157894736842, + "grad_norm": 0.71875, + "learning_rate": 7.742150468936647e-05, + "loss": 0.7358, + "step": 547 + }, + { + "epoch": 2.3073684210526317, + "grad_norm": 0.380859375, + "learning_rate": 7.709535604722684e-05, + "loss": 1.0316, + "step": 548 + }, + { + "epoch": 2.311578947368421, + "grad_norm": 0.64453125, + "learning_rate": 7.67694643365017e-05, + "loss": 0.9351, + "step": 549 + }, + { + "epoch": 2.3157894736842106, + "grad_norm": 0.498046875, + "learning_rate": 7.644383321286094e-05, + "loss": 1.0617, + "step": 550 + }, + { + "epoch": 2.32, + "grad_norm": 0.9140625, + "learning_rate": 7.61184663290514e-05, + "loss": 0.6994, + "step": 551 + }, + { + "epoch": 2.3242105263157895, + "grad_norm": 0.375, + "learning_rate": 7.579336733485584e-05, + "loss": 1.0248, + "step": 552 + }, + { + "epoch": 2.3284210526315787, + "grad_norm": 0.58984375, + "learning_rate": 7.546853987705191e-05, + "loss": 0.9696, + "step": 553 + }, + { + "epoch": 2.3326315789473684, + "grad_norm": 0.7265625, + "learning_rate": 7.514398759937135e-05, + "loss": 0.7576, + "step": 554 + }, + { + "epoch": 2.336842105263158, + "grad_norm": 0.369140625, + "learning_rate": 7.481971414245907e-05, + "loss": 0.7026, + "step": 555 + }, + { + "epoch": 2.3410526315789473, + "grad_norm": 0.69921875, + "learning_rate": 7.449572314383237e-05, + "loss": 0.8625, + "step": 556 + }, + { + "epoch": 2.345263157894737, + "grad_norm": 0.921875, + "learning_rate": 7.417201823784005e-05, + "loss": 0.849, + "step": 557 + }, + { + "epoch": 2.349473684210526, + "grad_norm": 0.453125, + "learning_rate": 7.384860305562172e-05, + "loss": 1.274, + "step": 558 + }, + { + "epoch": 2.353684210526316, + "grad_norm": 0.73828125, + "learning_rate": 7.35254812250669e-05, + "loss": 1.0279, + "step": 559 + }, + { + "epoch": 2.3578947368421055, + "grad_norm": 0.353515625, + "learning_rate": 7.320265637077473e-05, + "loss": 1.0411, + "step": 560 + }, + { + "epoch": 2.3621052631578947, + "grad_norm": 0.7109375, + "learning_rate": 7.288013211401278e-05, + "loss": 1.1542, + "step": 561 + }, + { + "epoch": 2.3663157894736844, + "grad_norm": 0.55078125, + "learning_rate": 7.255791207267679e-05, + "loss": 1.1941, + "step": 562 + }, + { + "epoch": 2.3705263157894736, + "grad_norm": 0.58203125, + "learning_rate": 7.223599986124994e-05, + "loss": 1.0699, + "step": 563 + }, + { + "epoch": 2.3747368421052633, + "grad_norm": 0.76953125, + "learning_rate": 7.191439909076243e-05, + "loss": 1.0074, + "step": 564 + }, + { + "epoch": 2.3789473684210525, + "grad_norm": 0.408203125, + "learning_rate": 7.159311336875083e-05, + "loss": 1.1968, + "step": 565 + }, + { + "epoch": 2.383157894736842, + "grad_norm": 1.0234375, + "learning_rate": 7.127214629921765e-05, + "loss": 0.9567, + "step": 566 + }, + { + "epoch": 2.3873684210526314, + "grad_norm": 0.5, + "learning_rate": 7.095150148259093e-05, + "loss": 0.585, + "step": 567 + }, + { + "epoch": 2.391578947368421, + "grad_norm": 0.52734375, + "learning_rate": 7.0631182515684e-05, + "loss": 0.9657, + "step": 568 + }, + { + "epoch": 2.3957894736842107, + "grad_norm": 0.43359375, + "learning_rate": 7.031119299165483e-05, + "loss": 0.7711, + "step": 569 + }, + { + "epoch": 2.4, + "grad_norm": 0.43359375, + "learning_rate": 6.999153649996595e-05, + "loss": 0.9455, + "step": 570 + }, + { + "epoch": 2.4042105263157896, + "grad_norm": 0.6953125, + "learning_rate": 6.967221662634407e-05, + "loss": 0.9143, + "step": 571 + }, + { + "epoch": 2.408421052631579, + "grad_norm": 0.443359375, + "learning_rate": 6.935323695274002e-05, + "loss": 1.3066, + "step": 572 + }, + { + "epoch": 2.4126315789473685, + "grad_norm": 0.4140625, + "learning_rate": 6.903460105728834e-05, + "loss": 1.02, + "step": 573 + }, + { + "epoch": 2.416842105263158, + "grad_norm": 0.322265625, + "learning_rate": 6.871631251426728e-05, + "loss": 1.0387, + "step": 574 + }, + { + "epoch": 2.4210526315789473, + "grad_norm": 0.625, + "learning_rate": 6.839837489405881e-05, + "loss": 0.9289, + "step": 575 + }, + { + "epoch": 2.425263157894737, + "grad_norm": 0.34765625, + "learning_rate": 6.808079176310827e-05, + "loss": 1.0434, + "step": 576 + }, + { + "epoch": 2.4294736842105262, + "grad_norm": 0.69921875, + "learning_rate": 6.776356668388464e-05, + "loss": 0.9514, + "step": 577 + }, + { + "epoch": 2.433684210526316, + "grad_norm": 0.486328125, + "learning_rate": 6.744670321484043e-05, + "loss": 0.8505, + "step": 578 + }, + { + "epoch": 2.437894736842105, + "grad_norm": 0.447265625, + "learning_rate": 6.713020491037188e-05, + "loss": 1.2355, + "step": 579 + }, + { + "epoch": 2.442105263157895, + "grad_norm": 0.359375, + "learning_rate": 6.681407532077895e-05, + "loss": 0.9406, + "step": 580 + }, + { + "epoch": 2.446315789473684, + "grad_norm": 0.6953125, + "learning_rate": 6.649831799222557e-05, + "loss": 0.8115, + "step": 581 + }, + { + "epoch": 2.4505263157894737, + "grad_norm": 0.5390625, + "learning_rate": 6.618293646669986e-05, + "loss": 1.0947, + "step": 582 + }, + { + "epoch": 2.4547368421052633, + "grad_norm": 0.69921875, + "learning_rate": 6.586793428197447e-05, + "loss": 1.1178, + "step": 583 + }, + { + "epoch": 2.4589473684210525, + "grad_norm": 0.427734375, + "learning_rate": 6.555331497156672e-05, + "loss": 0.8136, + "step": 584 + }, + { + "epoch": 2.463157894736842, + "grad_norm": 0.87109375, + "learning_rate": 6.523908206469909e-05, + "loss": 1.0041, + "step": 585 + }, + { + "epoch": 2.4673684210526314, + "grad_norm": 0.4375, + "learning_rate": 6.492523908625959e-05, + "loss": 1.1051, + "step": 586 + }, + { + "epoch": 2.471578947368421, + "grad_norm": 0.5625, + "learning_rate": 6.461178955676228e-05, + "loss": 1.0417, + "step": 587 + }, + { + "epoch": 2.4757894736842108, + "grad_norm": 0.35546875, + "learning_rate": 6.42987369923077e-05, + "loss": 0.9521, + "step": 588 + }, + { + "epoch": 2.48, + "grad_norm": 0.375, + "learning_rate": 6.398608490454346e-05, + "loss": 1.3175, + "step": 589 + }, + { + "epoch": 2.4842105263157896, + "grad_norm": 0.40625, + "learning_rate": 6.367383680062478e-05, + "loss": 1.0413, + "step": 590 + }, + { + "epoch": 2.488421052631579, + "grad_norm": 0.625, + "learning_rate": 6.336199618317537e-05, + "loss": 0.8431, + "step": 591 + }, + { + "epoch": 2.4926315789473685, + "grad_norm": 0.451171875, + "learning_rate": 6.30505665502479e-05, + "loss": 0.8335, + "step": 592 + }, + { + "epoch": 2.4968421052631578, + "grad_norm": 0.83203125, + "learning_rate": 6.27395513952848e-05, + "loss": 0.689, + "step": 593 + }, + { + "epoch": 2.5010526315789474, + "grad_norm": 0.3984375, + "learning_rate": 6.242895420707917e-05, + "loss": 1.1857, + "step": 594 + }, + { + "epoch": 2.5052631578947366, + "grad_norm": 0.375, + "learning_rate": 6.211877846973563e-05, + "loss": 0.6606, + "step": 595 + }, + { + "epoch": 2.5094736842105263, + "grad_norm": 0.578125, + "learning_rate": 6.180902766263113e-05, + "loss": 0.892, + "step": 596 + }, + { + "epoch": 2.513684210526316, + "grad_norm": 0.427734375, + "learning_rate": 6.149970526037603e-05, + "loss": 1.0731, + "step": 597 + }, + { + "epoch": 2.517894736842105, + "grad_norm": 0.578125, + "learning_rate": 6.119081473277501e-05, + "loss": 1.0982, + "step": 598 + }, + { + "epoch": 2.522105263157895, + "grad_norm": 0.376953125, + "learning_rate": 6.088235954478842e-05, + "loss": 1.0339, + "step": 599 + }, + { + "epoch": 2.526315789473684, + "grad_norm": 0.384765625, + "learning_rate": 6.057434315649304e-05, + "loss": 1.1655, + "step": 600 + }, + { + "epoch": 2.526315789473684, + "eval_loss": 1.217448353767395, + "eval_runtime": 25.9078, + "eval_samples_per_second": 3.86, + "eval_steps_per_second": 1.93, + "step": 600 + }, + { + "epoch": 2.5305263157894737, + "grad_norm": 0.953125, + "learning_rate": 6.026676902304345e-05, + "loss": 0.8561, + "step": 601 + }, + { + "epoch": 2.5347368421052634, + "grad_norm": 0.625, + "learning_rate": 5.99596405946333e-05, + "loss": 0.8852, + "step": 602 + }, + { + "epoch": 2.5389473684210526, + "grad_norm": 0.423828125, + "learning_rate": 5.965296131645663e-05, + "loss": 0.8336, + "step": 603 + }, + { + "epoch": 2.543157894736842, + "grad_norm": 1.3515625, + "learning_rate": 5.9346734628669065e-05, + "loss": 0.6848, + "step": 604 + }, + { + "epoch": 2.5473684210526315, + "grad_norm": 0.8046875, + "learning_rate": 5.904096396634935e-05, + "loss": 0.951, + "step": 605 + }, + { + "epoch": 2.551578947368421, + "grad_norm": 0.44921875, + "learning_rate": 5.873565275946088e-05, + "loss": 0.877, + "step": 606 + }, + { + "epoch": 2.5557894736842104, + "grad_norm": 0.60546875, + "learning_rate": 5.843080443281303e-05, + "loss": 0.9183, + "step": 607 + }, + { + "epoch": 2.56, + "grad_norm": 0.71484375, + "learning_rate": 5.8126422406022885e-05, + "loss": 0.6659, + "step": 608 + }, + { + "epoch": 2.5642105263157893, + "grad_norm": 0.54296875, + "learning_rate": 5.782251009347679e-05, + "loss": 1.0883, + "step": 609 + }, + { + "epoch": 2.568421052631579, + "grad_norm": 0.453125, + "learning_rate": 5.7519070904292247e-05, + "loss": 0.963, + "step": 610 + }, + { + "epoch": 2.5726315789473686, + "grad_norm": 0.43359375, + "learning_rate": 5.7216108242279386e-05, + "loss": 1.2451, + "step": 611 + }, + { + "epoch": 2.576842105263158, + "grad_norm": 0.78515625, + "learning_rate": 5.691362550590297e-05, + "loss": 0.8804, + "step": 612 + }, + { + "epoch": 2.5810526315789475, + "grad_norm": 0.8515625, + "learning_rate": 5.6611626088244194e-05, + "loss": 1.0411, + "step": 613 + }, + { + "epoch": 2.5852631578947367, + "grad_norm": 0.42578125, + "learning_rate": 5.631011337696271e-05, + "loss": 0.9097, + "step": 614 + }, + { + "epoch": 2.5894736842105264, + "grad_norm": 0.478515625, + "learning_rate": 5.6009090754258545e-05, + "loss": 1.0427, + "step": 615 + }, + { + "epoch": 2.593684210526316, + "grad_norm": 0.40625, + "learning_rate": 5.570856159683418e-05, + "loss": 0.8898, + "step": 616 + }, + { + "epoch": 2.5978947368421053, + "grad_norm": 0.373046875, + "learning_rate": 5.5408529275856624e-05, + "loss": 0.9602, + "step": 617 + }, + { + "epoch": 2.6021052631578945, + "grad_norm": 0.5390625, + "learning_rate": 5.510899715691984e-05, + "loss": 1.0589, + "step": 618 + }, + { + "epoch": 2.606315789473684, + "grad_norm": 0.4765625, + "learning_rate": 5.4809968600006635e-05, + "loss": 0.939, + "step": 619 + }, + { + "epoch": 2.610526315789474, + "grad_norm": 0.44921875, + "learning_rate": 5.451144695945116e-05, + "loss": 0.909, + "step": 620 + }, + { + "epoch": 2.614736842105263, + "grad_norm": 1.203125, + "learning_rate": 5.421343558390128e-05, + "loss": 1.0197, + "step": 621 + }, + { + "epoch": 2.6189473684210527, + "grad_norm": 0.96875, + "learning_rate": 5.3915937816281095e-05, + "loss": 0.991, + "step": 622 + }, + { + "epoch": 2.623157894736842, + "grad_norm": 0.85546875, + "learning_rate": 5.361895699375319e-05, + "loss": 0.9968, + "step": 623 + }, + { + "epoch": 2.6273684210526316, + "grad_norm": 0.890625, + "learning_rate": 5.3322496447681414e-05, + "loss": 1.0106, + "step": 624 + }, + { + "epoch": 2.6315789473684212, + "grad_norm": 0.41015625, + "learning_rate": 5.302655950359341e-05, + "loss": 1.0636, + "step": 625 + }, + { + "epoch": 2.6357894736842105, + "grad_norm": 0.44921875, + "learning_rate": 5.273114948114346e-05, + "loss": 0.7656, + "step": 626 + }, + { + "epoch": 2.64, + "grad_norm": 0.5078125, + "learning_rate": 5.2436269694075046e-05, + "loss": 1.1255, + "step": 627 + }, + { + "epoch": 2.6442105263157893, + "grad_norm": 2.515625, + "learning_rate": 5.214192345018374e-05, + "loss": 0.8405, + "step": 628 + }, + { + "epoch": 2.648421052631579, + "grad_norm": 0.375, + "learning_rate": 5.1848114051280195e-05, + "loss": 1.0719, + "step": 629 + }, + { + "epoch": 2.6526315789473687, + "grad_norm": 0.353515625, + "learning_rate": 5.1554844793153e-05, + "loss": 0.9582, + "step": 630 + }, + { + "epoch": 2.656842105263158, + "grad_norm": 0.46875, + "learning_rate": 5.126211896553178e-05, + "loss": 0.8587, + "step": 631 + }, + { + "epoch": 2.661052631578947, + "grad_norm": 0.8203125, + "learning_rate": 5.096993985205023e-05, + "loss": 1.055, + "step": 632 + }, + { + "epoch": 2.665263157894737, + "grad_norm": 0.48046875, + "learning_rate": 5.0678310730209275e-05, + "loss": 1.0103, + "step": 633 + }, + { + "epoch": 2.6694736842105264, + "grad_norm": 0.70703125, + "learning_rate": 5.0387234871340486e-05, + "loss": 0.7718, + "step": 634 + }, + { + "epoch": 2.6736842105263157, + "grad_norm": 0.72265625, + "learning_rate": 5.009671554056911e-05, + "loss": 0.9338, + "step": 635 + }, + { + "epoch": 2.6778947368421053, + "grad_norm": 0.373046875, + "learning_rate": 4.980675599677757e-05, + "loss": 0.7023, + "step": 636 + }, + { + "epoch": 2.6821052631578945, + "grad_norm": 0.66796875, + "learning_rate": 4.9517359492569035e-05, + "loss": 1.1044, + "step": 637 + }, + { + "epoch": 2.686315789473684, + "grad_norm": 0.5234375, + "learning_rate": 4.9228529274230695e-05, + "loss": 0.9719, + "step": 638 + }, + { + "epoch": 2.690526315789474, + "grad_norm": 0.55078125, + "learning_rate": 4.894026858169752e-05, + "loss": 1.1118, + "step": 639 + }, + { + "epoch": 2.694736842105263, + "grad_norm": 0.490234375, + "learning_rate": 4.865258064851579e-05, + "loss": 1.2393, + "step": 640 + }, + { + "epoch": 2.6989473684210528, + "grad_norm": 0.443359375, + "learning_rate": 4.8365468701807016e-05, + "loss": 0.8362, + "step": 641 + }, + { + "epoch": 2.703157894736842, + "grad_norm": 0.82421875, + "learning_rate": 4.807893596223152e-05, + "loss": 0.9748, + "step": 642 + }, + { + "epoch": 2.7073684210526316, + "grad_norm": 0.6328125, + "learning_rate": 4.7792985643952435e-05, + "loss": 1.1939, + "step": 643 + }, + { + "epoch": 2.7115789473684213, + "grad_norm": 0.51171875, + "learning_rate": 4.75076209545996e-05, + "loss": 0.8513, + "step": 644 + }, + { + "epoch": 2.7157894736842105, + "grad_norm": 0.47265625, + "learning_rate": 4.722284509523362e-05, + "loss": 0.8377, + "step": 645 + }, + { + "epoch": 2.7199999999999998, + "grad_norm": 0.7109375, + "learning_rate": 4.693866126030995e-05, + "loss": 1.0126, + "step": 646 + }, + { + "epoch": 2.7242105263157894, + "grad_norm": 0.57421875, + "learning_rate": 4.665507263764299e-05, + "loss": 0.6115, + "step": 647 + }, + { + "epoch": 2.728421052631579, + "grad_norm": 0.88671875, + "learning_rate": 4.637208240837042e-05, + "loss": 1.0717, + "step": 648 + }, + { + "epoch": 2.7326315789473683, + "grad_norm": 0.56640625, + "learning_rate": 4.608969374691753e-05, + "loss": 1.1582, + "step": 649 + }, + { + "epoch": 2.736842105263158, + "grad_norm": 0.91796875, + "learning_rate": 4.5807909820961494e-05, + "loss": 1.1218, + "step": 650 + }, + { + "epoch": 2.741052631578947, + "grad_norm": 0.6328125, + "learning_rate": 4.5526733791395896e-05, + "loss": 0.9114, + "step": 651 + }, + { + "epoch": 2.745263157894737, + "grad_norm": 0.70703125, + "learning_rate": 4.5246168812295286e-05, + "loss": 0.957, + "step": 652 + }, + { + "epoch": 2.7494736842105265, + "grad_norm": 0.5859375, + "learning_rate": 4.496621803087988e-05, + "loss": 0.6728, + "step": 653 + }, + { + "epoch": 2.7536842105263157, + "grad_norm": 1.2109375, + "learning_rate": 4.468688458748006e-05, + "loss": 0.8425, + "step": 654 + }, + { + "epoch": 2.7578947368421054, + "grad_norm": 0.5703125, + "learning_rate": 4.4408171615501316e-05, + "loss": 0.6478, + "step": 655 + }, + { + "epoch": 2.7621052631578946, + "grad_norm": 0.28125, + "learning_rate": 4.413008224138897e-05, + "loss": 0.9384, + "step": 656 + }, + { + "epoch": 2.7663157894736843, + "grad_norm": 0.734375, + "learning_rate": 4.38526195845933e-05, + "loss": 0.9842, + "step": 657 + }, + { + "epoch": 2.770526315789474, + "grad_norm": 0.8125, + "learning_rate": 4.357578675753432e-05, + "loss": 0.8117, + "step": 658 + }, + { + "epoch": 2.774736842105263, + "grad_norm": 0.6328125, + "learning_rate": 4.329958686556699e-05, + "loss": 0.8619, + "step": 659 + }, + { + "epoch": 2.7789473684210524, + "grad_norm": 0.55859375, + "learning_rate": 4.302402300694636e-05, + "loss": 1.1503, + "step": 660 + }, + { + "epoch": 2.7789473684210524, + "eval_loss": 1.2197896242141724, + "eval_runtime": 25.8587, + "eval_samples_per_second": 3.867, + "eval_steps_per_second": 1.934, + "step": 660 + }, + { + "epoch": 2.783157894736842, + "grad_norm": 0.59765625, + "learning_rate": 4.274909827279283e-05, + "loss": 0.7691, + "step": 661 + }, + { + "epoch": 2.7873684210526317, + "grad_norm": 0.54296875, + "learning_rate": 4.247481574705744e-05, + "loss": 0.5837, + "step": 662 + }, + { + "epoch": 2.791578947368421, + "grad_norm": 0.80078125, + "learning_rate": 4.220117850648733e-05, + "loss": 0.8718, + "step": 663 + }, + { + "epoch": 2.7957894736842106, + "grad_norm": 0.75390625, + "learning_rate": 4.1928189620591116e-05, + "loss": 0.8497, + "step": 664 + }, + { + "epoch": 2.8, + "grad_norm": 0.359375, + "learning_rate": 4.165585215160468e-05, + "loss": 1.1803, + "step": 665 + }, + { + "epoch": 2.8042105263157895, + "grad_norm": 0.78515625, + "learning_rate": 4.138416915445655e-05, + "loss": 0.7498, + "step": 666 + }, + { + "epoch": 2.808421052631579, + "grad_norm": 0.451171875, + "learning_rate": 4.1113143676733757e-05, + "loss": 1.2798, + "step": 667 + }, + { + "epoch": 2.8126315789473684, + "grad_norm": 0.341796875, + "learning_rate": 4.084277875864776e-05, + "loss": 0.9448, + "step": 668 + }, + { + "epoch": 2.816842105263158, + "grad_norm": 0.625, + "learning_rate": 4.0573077433000115e-05, + "loss": 1.0791, + "step": 669 + }, + { + "epoch": 2.8210526315789473, + "grad_norm": 0.3671875, + "learning_rate": 4.030404272514864e-05, + "loss": 0.9928, + "step": 670 + }, + { + "epoch": 2.825263157894737, + "grad_norm": 0.6015625, + "learning_rate": 4.003567765297331e-05, + "loss": 1.0038, + "step": 671 + }, + { + "epoch": 2.8294736842105266, + "grad_norm": 0.46875, + "learning_rate": 3.9767985226842696e-05, + "loss": 1.294, + "step": 672 + }, + { + "epoch": 2.833684210526316, + "grad_norm": 0.34765625, + "learning_rate": 3.950096844957981e-05, + "loss": 1.1689, + "step": 673 + }, + { + "epoch": 2.837894736842105, + "grad_norm": 0.609375, + "learning_rate": 3.923463031642872e-05, + "loss": 1.2769, + "step": 674 + }, + { + "epoch": 2.8421052631578947, + "grad_norm": 0.43359375, + "learning_rate": 3.8968973815020806e-05, + "loss": 1.0042, + "step": 675 + }, + { + "epoch": 2.8463157894736844, + "grad_norm": 0.58984375, + "learning_rate": 3.870400192534128e-05, + "loss": 0.8091, + "step": 676 + }, + { + "epoch": 2.8505263157894736, + "grad_norm": 0.5, + "learning_rate": 3.843971761969583e-05, + "loss": 0.9447, + "step": 677 + }, + { + "epoch": 2.8547368421052632, + "grad_norm": 0.58984375, + "learning_rate": 3.81761238626771e-05, + "loss": 0.7497, + "step": 678 + }, + { + "epoch": 2.8589473684210525, + "grad_norm": 0.66015625, + "learning_rate": 3.791322361113164e-05, + "loss": 0.7618, + "step": 679 + }, + { + "epoch": 2.863157894736842, + "grad_norm": 0.5078125, + "learning_rate": 3.7651019814126654e-05, + "loss": 1.1192, + "step": 680 + }, + { + "epoch": 2.867368421052632, + "grad_norm": 0.484375, + "learning_rate": 3.738951541291685e-05, + "loss": 0.8821, + "step": 681 + }, + { + "epoch": 2.871578947368421, + "grad_norm": 0.625, + "learning_rate": 3.7128713340911535e-05, + "loss": 0.868, + "step": 682 + }, + { + "epoch": 2.8757894736842107, + "grad_norm": 0.328125, + "learning_rate": 3.686861652364164e-05, + "loss": 1.0257, + "step": 683 + }, + { + "epoch": 2.88, + "grad_norm": 0.376953125, + "learning_rate": 3.660922787872706e-05, + "loss": 0.8358, + "step": 684 + }, + { + "epoch": 2.8842105263157896, + "grad_norm": 0.470703125, + "learning_rate": 3.635055031584369e-05, + "loss": 1.0488, + "step": 685 + }, + { + "epoch": 2.8884210526315792, + "grad_norm": 0.4140625, + "learning_rate": 3.609258673669097e-05, + "loss": 1.0167, + "step": 686 + }, + { + "epoch": 2.8926315789473684, + "grad_norm": 0.44140625, + "learning_rate": 3.58353400349592e-05, + "loss": 1.1384, + "step": 687 + }, + { + "epoch": 2.8968421052631577, + "grad_norm": 0.69921875, + "learning_rate": 3.557881309629729e-05, + "loss": 0.847, + "step": 688 + }, + { + "epoch": 2.9010526315789473, + "grad_norm": 1.0703125, + "learning_rate": 3.532300879828013e-05, + "loss": 0.9456, + "step": 689 + }, + { + "epoch": 2.905263157894737, + "grad_norm": 0.890625, + "learning_rate": 3.5067930010376484e-05, + "loss": 0.8495, + "step": 690 + }, + { + "epoch": 2.909473684210526, + "grad_norm": 0.91015625, + "learning_rate": 3.481357959391671e-05, + "loss": 1.0274, + "step": 691 + }, + { + "epoch": 2.913684210526316, + "grad_norm": 0.4765625, + "learning_rate": 3.455996040206076e-05, + "loss": 1.0327, + "step": 692 + }, + { + "epoch": 2.917894736842105, + "grad_norm": 0.64453125, + "learning_rate": 3.43070752797661e-05, + "loss": 0.83, + "step": 693 + }, + { + "epoch": 2.9221052631578948, + "grad_norm": 0.8671875, + "learning_rate": 3.4054927063755796e-05, + "loss": 0.786, + "step": 694 + }, + { + "epoch": 2.9263157894736844, + "grad_norm": 0.67578125, + "learning_rate": 3.38035185824867e-05, + "loss": 0.8956, + "step": 695 + }, + { + "epoch": 2.9305263157894736, + "grad_norm": 0.61328125, + "learning_rate": 3.355285265611784e-05, + "loss": 0.8357, + "step": 696 + }, + { + "epoch": 2.9347368421052633, + "grad_norm": 0.498046875, + "learning_rate": 3.3302932096478554e-05, + "loss": 1.0673, + "step": 697 + }, + { + "epoch": 2.9389473684210525, + "grad_norm": 0.53515625, + "learning_rate": 3.305375970703711e-05, + "loss": 0.8016, + "step": 698 + }, + { + "epoch": 2.943157894736842, + "grad_norm": 0.62890625, + "learning_rate": 3.28053382828693e-05, + "loss": 0.817, + "step": 699 + }, + { + "epoch": 2.9473684210526314, + "grad_norm": 0.62890625, + "learning_rate": 3.2557670610626925e-05, + "loss": 1.0499, + "step": 700 + }, + { + "epoch": 2.951578947368421, + "grad_norm": 0.9375, + "learning_rate": 3.231075946850663e-05, + "loss": 0.8511, + "step": 701 + }, + { + "epoch": 2.9557894736842103, + "grad_norm": 0.455078125, + "learning_rate": 3.206460762621873e-05, + "loss": 1.0486, + "step": 702 + }, + { + "epoch": 2.96, + "grad_norm": 0.38671875, + "learning_rate": 3.1819217844956214e-05, + "loss": 0.9743, + "step": 703 + }, + { + "epoch": 2.9642105263157896, + "grad_norm": 0.42578125, + "learning_rate": 3.157459287736362e-05, + "loss": 0.9247, + "step": 704 + }, + { + "epoch": 2.968421052631579, + "grad_norm": 0.3359375, + "learning_rate": 3.1330735467506246e-05, + "loss": 1.0188, + "step": 705 + }, + { + "epoch": 2.9726315789473685, + "grad_norm": 0.51953125, + "learning_rate": 3.108764835083938e-05, + "loss": 1.1879, + "step": 706 + }, + { + "epoch": 2.9768421052631577, + "grad_norm": 0.41015625, + "learning_rate": 3.084533425417758e-05, + "loss": 0.6776, + "step": 707 + }, + { + "epoch": 2.9810526315789474, + "grad_norm": 0.349609375, + "learning_rate": 3.0603795895664124e-05, + "loss": 0.9761, + "step": 708 + }, + { + "epoch": 2.985263157894737, + "grad_norm": 0.7109375, + "learning_rate": 3.036303598474045e-05, + "loss": 0.9003, + "step": 709 + }, + { + "epoch": 2.9894736842105263, + "grad_norm": 0.38671875, + "learning_rate": 3.0123057222115836e-05, + "loss": 1.0767, + "step": 710 + }, + { + "epoch": 2.993684210526316, + "grad_norm": 0.52734375, + "learning_rate": 2.988386229973713e-05, + "loss": 0.792, + "step": 711 + }, + { + "epoch": 2.997894736842105, + "grad_norm": 0.5078125, + "learning_rate": 2.964545390075841e-05, + "loss": 1.0222, + "step": 712 + }, + { + "epoch": 3.002105263157895, + "grad_norm": 0.453125, + "learning_rate": 2.940783469951103e-05, + "loss": 0.926, + "step": 713 + }, + { + "epoch": 3.006315789473684, + "grad_norm": 0.486328125, + "learning_rate": 2.9171007361473514e-05, + "loss": 1.0828, + "step": 714 + }, + { + "epoch": 3.0105263157894737, + "grad_norm": 0.77734375, + "learning_rate": 2.893497454324182e-05, + "loss": 0.9486, + "step": 715 + }, + { + "epoch": 3.014736842105263, + "grad_norm": 0.7734375, + "learning_rate": 2.8699738892499328e-05, + "loss": 0.6997, + "step": 716 + }, + { + "epoch": 3.0189473684210526, + "grad_norm": 0.515625, + "learning_rate": 2.8465303047987267e-05, + "loss": 1.1117, + "step": 717 + }, + { + "epoch": 3.0231578947368423, + "grad_norm": 0.765625, + "learning_rate": 2.8231669639475067e-05, + "loss": 1.03, + "step": 718 + }, + { + "epoch": 3.0273684210526315, + "grad_norm": 0.46484375, + "learning_rate": 2.799884128773097e-05, + "loss": 0.9977, + "step": 719 + }, + { + "epoch": 3.031578947368421, + "grad_norm": 0.384765625, + "learning_rate": 2.776682060449247e-05, + "loss": 0.9577, + "step": 720 + }, + { + "epoch": 3.031578947368421, + "eval_loss": 1.2163540124893188, + "eval_runtime": 25.8624, + "eval_samples_per_second": 3.867, + "eval_steps_per_second": 1.933, + "step": 720 + }, + { + "epoch": 3.0357894736842104, + "grad_norm": 0.357421875, + "learning_rate": 2.753561019243709e-05, + "loss": 0.9971, + "step": 721 + }, + { + "epoch": 3.04, + "grad_norm": 0.39453125, + "learning_rate": 2.7305212645153212e-05, + "loss": 1.1053, + "step": 722 + }, + { + "epoch": 3.0442105263157893, + "grad_norm": 0.322265625, + "learning_rate": 2.707563054711093e-05, + "loss": 0.8554, + "step": 723 + }, + { + "epoch": 3.048421052631579, + "grad_norm": 0.51171875, + "learning_rate": 2.6846866473633125e-05, + "loss": 0.8671, + "step": 724 + }, + { + "epoch": 3.0526315789473686, + "grad_norm": 0.388671875, + "learning_rate": 2.66189229908665e-05, + "loss": 0.5801, + "step": 725 + }, + { + "epoch": 3.056842105263158, + "grad_norm": 0.55859375, + "learning_rate": 2.6391802655752853e-05, + "loss": 0.9976, + "step": 726 + }, + { + "epoch": 3.0610526315789475, + "grad_norm": 0.447265625, + "learning_rate": 2.6165508016000385e-05, + "loss": 0.8534, + "step": 727 + }, + { + "epoch": 3.0652631578947367, + "grad_norm": 0.6796875, + "learning_rate": 2.594004161005511e-05, + "loss": 0.7658, + "step": 728 + }, + { + "epoch": 3.0694736842105264, + "grad_norm": 0.4296875, + "learning_rate": 2.5715405967072347e-05, + "loss": 0.8652, + "step": 729 + }, + { + "epoch": 3.0736842105263156, + "grad_norm": 0.515625, + "learning_rate": 2.549160360688838e-05, + "loss": 0.9121, + "step": 730 + }, + { + "epoch": 3.0778947368421052, + "grad_norm": 0.66015625, + "learning_rate": 2.5268637039992293e-05, + "loss": 0.9808, + "step": 731 + }, + { + "epoch": 3.082105263157895, + "grad_norm": 0.35546875, + "learning_rate": 2.50465087674976e-05, + "loss": 0.8625, + "step": 732 + }, + { + "epoch": 3.086315789473684, + "grad_norm": 0.55078125, + "learning_rate": 2.4825221281114296e-05, + "loss": 0.723, + "step": 733 + }, + { + "epoch": 3.090526315789474, + "grad_norm": 0.6171875, + "learning_rate": 2.4604777063121033e-05, + "loss": 0.9536, + "step": 734 + }, + { + "epoch": 3.094736842105263, + "grad_norm": 0.37890625, + "learning_rate": 2.438517858633703e-05, + "loss": 0.9585, + "step": 735 + }, + { + "epoch": 3.0989473684210527, + "grad_norm": 0.50390625, + "learning_rate": 2.4166428314094514e-05, + "loss": 0.4804, + "step": 736 + }, + { + "epoch": 3.103157894736842, + "grad_norm": 0.396484375, + "learning_rate": 2.3948528700211003e-05, + "loss": 1.0552, + "step": 737 + }, + { + "epoch": 3.1073684210526316, + "grad_norm": 0.337890625, + "learning_rate": 2.3731482188961818e-05, + "loss": 0.9594, + "step": 738 + }, + { + "epoch": 3.111578947368421, + "grad_norm": 0.38671875, + "learning_rate": 2.351529121505266e-05, + "loss": 0.9773, + "step": 739 + }, + { + "epoch": 3.1157894736842104, + "grad_norm": 0.46484375, + "learning_rate": 2.32999582035923e-05, + "loss": 0.7401, + "step": 740 + }, + { + "epoch": 3.12, + "grad_norm": 0.4453125, + "learning_rate": 2.3085485570065313e-05, + "loss": 0.9351, + "step": 741 + }, + { + "epoch": 3.1242105263157893, + "grad_norm": 0.8203125, + "learning_rate": 2.287187572030516e-05, + "loss": 0.7592, + "step": 742 + }, + { + "epoch": 3.128421052631579, + "grad_norm": 0.404296875, + "learning_rate": 2.2659131050466964e-05, + "loss": 1.0506, + "step": 743 + }, + { + "epoch": 3.132631578947368, + "grad_norm": 0.57421875, + "learning_rate": 2.244725394700079e-05, + "loss": 1.0695, + "step": 744 + }, + { + "epoch": 3.136842105263158, + "grad_norm": 0.60546875, + "learning_rate": 2.2236246786624792e-05, + "loss": 0.914, + "step": 745 + }, + { + "epoch": 3.1410526315789475, + "grad_norm": 0.455078125, + "learning_rate": 2.202611193629869e-05, + "loss": 0.7936, + "step": 746 + }, + { + "epoch": 3.1452631578947368, + "grad_norm": 0.5546875, + "learning_rate": 2.181685175319702e-05, + "loss": 0.7292, + "step": 747 + }, + { + "epoch": 3.1494736842105264, + "grad_norm": 0.671875, + "learning_rate": 2.160846858468285e-05, + "loss": 0.7791, + "step": 748 + }, + { + "epoch": 3.1536842105263156, + "grad_norm": 0.58203125, + "learning_rate": 2.1400964768281318e-05, + "loss": 1.0173, + "step": 749 + }, + { + "epoch": 3.1578947368421053, + "grad_norm": 0.60546875, + "learning_rate": 2.1194342631653607e-05, + "loss": 0.9076, + "step": 750 + }, + { + "epoch": 3.1621052631578945, + "grad_norm": 0.62890625, + "learning_rate": 2.0988604492570606e-05, + "loss": 0.7836, + "step": 751 + }, + { + "epoch": 3.166315789473684, + "grad_norm": 0.5078125, + "learning_rate": 2.0783752658887066e-05, + "loss": 0.8939, + "step": 752 + }, + { + "epoch": 3.170526315789474, + "grad_norm": 0.4296875, + "learning_rate": 2.057978942851565e-05, + "loss": 0.9172, + "step": 753 + }, + { + "epoch": 3.174736842105263, + "grad_norm": 0.59765625, + "learning_rate": 2.0376717089401164e-05, + "loss": 0.7943, + "step": 754 + }, + { + "epoch": 3.1789473684210527, + "grad_norm": 0.4296875, + "learning_rate": 2.0174537919494897e-05, + "loss": 1.0665, + "step": 755 + }, + { + "epoch": 3.183157894736842, + "grad_norm": 0.50390625, + "learning_rate": 1.9973254186729086e-05, + "loss": 1.0992, + "step": 756 + }, + { + "epoch": 3.1873684210526316, + "grad_norm": 0.447265625, + "learning_rate": 1.9772868148991395e-05, + "loss": 0.7196, + "step": 757 + }, + { + "epoch": 3.191578947368421, + "grad_norm": 0.703125, + "learning_rate": 1.9573382054099786e-05, + "loss": 0.6875, + "step": 758 + }, + { + "epoch": 3.1957894736842105, + "grad_norm": 0.453125, + "learning_rate": 1.937479813977703e-05, + "loss": 0.8144, + "step": 759 + }, + { + "epoch": 3.2, + "grad_norm": 0.671875, + "learning_rate": 1.9177118633625814e-05, + "loss": 0.8838, + "step": 760 + }, + { + "epoch": 3.2042105263157894, + "grad_norm": 1.453125, + "learning_rate": 1.8980345753103612e-05, + "loss": 1.0451, + "step": 761 + }, + { + "epoch": 3.208421052631579, + "grad_norm": 0.34765625, + "learning_rate": 1.8784481705498015e-05, + "loss": 0.9438, + "step": 762 + }, + { + "epoch": 3.2126315789473683, + "grad_norm": 0.478515625, + "learning_rate": 1.8589528687901723e-05, + "loss": 1.0965, + "step": 763 + }, + { + "epoch": 3.216842105263158, + "grad_norm": 0.4765625, + "learning_rate": 1.8395488887188005e-05, + "loss": 0.9689, + "step": 764 + }, + { + "epoch": 3.221052631578947, + "grad_norm": 0.419921875, + "learning_rate": 1.82023644799863e-05, + "loss": 1.084, + "step": 765 + }, + { + "epoch": 3.225263157894737, + "grad_norm": 0.85546875, + "learning_rate": 1.8010157632657543e-05, + "loss": 0.5852, + "step": 766 + }, + { + "epoch": 3.2294736842105265, + "grad_norm": 0.703125, + "learning_rate": 1.7818870501270057e-05, + "loss": 0.9669, + "step": 767 + }, + { + "epoch": 3.2336842105263157, + "grad_norm": 0.828125, + "learning_rate": 1.762850523157532e-05, + "loss": 0.7386, + "step": 768 + }, + { + "epoch": 3.2378947368421054, + "grad_norm": 0.6875, + "learning_rate": 1.7439063958983858e-05, + "loss": 1.1098, + "step": 769 + }, + { + "epoch": 3.2421052631578946, + "grad_norm": 0.6328125, + "learning_rate": 1.7250548808541322e-05, + "loss": 0.9314, + "step": 770 + }, + { + "epoch": 3.2463157894736843, + "grad_norm": 0.48046875, + "learning_rate": 1.706296189490465e-05, + "loss": 0.8762, + "step": 771 + }, + { + "epoch": 3.2505263157894735, + "grad_norm": 0.45703125, + "learning_rate": 1.687630532231833e-05, + "loss": 0.9833, + "step": 772 + }, + { + "epoch": 3.254736842105263, + "grad_norm": 0.3359375, + "learning_rate": 1.6690581184590858e-05, + "loss": 1.0493, + "step": 773 + }, + { + "epoch": 3.258947368421053, + "grad_norm": 0.66796875, + "learning_rate": 1.6505791565071138e-05, + "loss": 0.6527, + "step": 774 + }, + { + "epoch": 3.263157894736842, + "grad_norm": 1.15625, + "learning_rate": 1.6321938536625192e-05, + "loss": 1.0197, + "step": 775 + }, + { + "epoch": 3.2673684210526317, + "grad_norm": 0.51953125, + "learning_rate": 1.613902416161288e-05, + "loss": 1.1179, + "step": 776 + }, + { + "epoch": 3.271578947368421, + "grad_norm": 0.5546875, + "learning_rate": 1.5957050491864843e-05, + "loss": 0.7595, + "step": 777 + }, + { + "epoch": 3.2757894736842106, + "grad_norm": 0.87109375, + "learning_rate": 1.5776019568659338e-05, + "loss": 0.9304, + "step": 778 + }, + { + "epoch": 3.2800000000000002, + "grad_norm": 0.5078125, + "learning_rate": 1.559593342269945e-05, + "loss": 0.9252, + "step": 779 + }, + { + "epoch": 3.2842105263157895, + "grad_norm": 0.44140625, + "learning_rate": 1.5416794074090258e-05, + "loss": 0.9943, + "step": 780 + }, + { + "epoch": 3.2842105263157895, + "eval_loss": 1.2286467552185059, + "eval_runtime": 25.9074, + "eval_samples_per_second": 3.86, + "eval_steps_per_second": 1.93, + "step": 780 + }, + { + "epoch": 3.288421052631579, + "grad_norm": 0.4921875, + "learning_rate": 1.5238603532316287e-05, + "loss": 0.8001, + "step": 781 + }, + { + "epoch": 3.2926315789473684, + "grad_norm": 0.91796875, + "learning_rate": 1.5061363796218785e-05, + "loss": 0.7387, + "step": 782 + }, + { + "epoch": 3.296842105263158, + "grad_norm": 0.609375, + "learning_rate": 1.4885076853973446e-05, + "loss": 1.1835, + "step": 783 + }, + { + "epoch": 3.3010526315789472, + "grad_norm": 0.451171875, + "learning_rate": 1.4709744683068039e-05, + "loss": 1.0573, + "step": 784 + }, + { + "epoch": 3.305263157894737, + "grad_norm": 0.42578125, + "learning_rate": 1.453536925028025e-05, + "loss": 0.9381, + "step": 785 + }, + { + "epoch": 3.309473684210526, + "grad_norm": 0.51953125, + "learning_rate": 1.4361952511655618e-05, + "loss": 1.0083, + "step": 786 + }, + { + "epoch": 3.313684210526316, + "grad_norm": 0.546875, + "learning_rate": 1.4189496412485592e-05, + "loss": 0.9212, + "step": 787 + }, + { + "epoch": 3.3178947368421055, + "grad_norm": 0.47265625, + "learning_rate": 1.4018002887285687e-05, + "loss": 0.936, + "step": 788 + }, + { + "epoch": 3.3221052631578947, + "grad_norm": 0.57421875, + "learning_rate": 1.3847473859773851e-05, + "loss": 0.8633, + "step": 789 + }, + { + "epoch": 3.3263157894736843, + "grad_norm": 0.6484375, + "learning_rate": 1.3677911242848806e-05, + "loss": 0.7159, + "step": 790 + }, + { + "epoch": 3.3305263157894736, + "grad_norm": 0.66796875, + "learning_rate": 1.3509316938568616e-05, + "loss": 0.8585, + "step": 791 + }, + { + "epoch": 3.334736842105263, + "grad_norm": 0.5859375, + "learning_rate": 1.334169283812936e-05, + "loss": 0.7944, + "step": 792 + }, + { + "epoch": 3.3389473684210524, + "grad_norm": 1.2421875, + "learning_rate": 1.3175040821843954e-05, + "loss": 0.7234, + "step": 793 + }, + { + "epoch": 3.343157894736842, + "grad_norm": 0.5546875, + "learning_rate": 1.300936275912098e-05, + "loss": 0.9608, + "step": 794 + }, + { + "epoch": 3.3473684210526318, + "grad_norm": 0.58984375, + "learning_rate": 1.284466050844374e-05, + "loss": 0.9962, + "step": 795 + }, + { + "epoch": 3.351578947368421, + "grad_norm": 0.453125, + "learning_rate": 1.2680935917349523e-05, + "loss": 1.0175, + "step": 796 + }, + { + "epoch": 3.3557894736842107, + "grad_norm": 0.515625, + "learning_rate": 1.251819082240866e-05, + "loss": 0.8974, + "step": 797 + }, + { + "epoch": 3.36, + "grad_norm": 0.6484375, + "learning_rate": 1.2356427049204122e-05, + "loss": 0.9817, + "step": 798 + }, + { + "epoch": 3.3642105263157895, + "grad_norm": 0.76953125, + "learning_rate": 1.2195646412310936e-05, + "loss": 0.915, + "step": 799 + }, + { + "epoch": 3.3684210526315788, + "grad_norm": 0.48046875, + "learning_rate": 1.2035850715275865e-05, + "loss": 0.7109, + "step": 800 + }, + { + "epoch": 3.3726315789473684, + "grad_norm": 0.4453125, + "learning_rate": 1.1877041750597173e-05, + "loss": 0.9533, + "step": 801 + }, + { + "epoch": 3.376842105263158, + "grad_norm": 0.51171875, + "learning_rate": 1.1719221299704497e-05, + "loss": 0.913, + "step": 802 + }, + { + "epoch": 3.3810526315789473, + "grad_norm": 0.38671875, + "learning_rate": 1.1562391132938899e-05, + "loss": 0.6807, + "step": 803 + }, + { + "epoch": 3.385263157894737, + "grad_norm": 0.53125, + "learning_rate": 1.1406553009533027e-05, + "loss": 0.7178, + "step": 804 + }, + { + "epoch": 3.389473684210526, + "grad_norm": 0.458984375, + "learning_rate": 1.1251708677591277e-05, + "loss": 0.9133, + "step": 805 + }, + { + "epoch": 3.393684210526316, + "grad_norm": 0.41015625, + "learning_rate": 1.1097859874070294e-05, + "loss": 1.0886, + "step": 806 + }, + { + "epoch": 3.397894736842105, + "grad_norm": 0.40625, + "learning_rate": 1.0945008324759398e-05, + "loss": 0.6535, + "step": 807 + }, + { + "epoch": 3.4021052631578947, + "grad_norm": 0.53515625, + "learning_rate": 1.0793155744261351e-05, + "loss": 0.7147, + "step": 808 + }, + { + "epoch": 3.406315789473684, + "grad_norm": 0.546875, + "learning_rate": 1.0642303835972989e-05, + "loss": 1.0274, + "step": 809 + }, + { + "epoch": 3.4105263157894736, + "grad_norm": 0.54296875, + "learning_rate": 1.0492454292066178e-05, + "loss": 0.9565, + "step": 810 + }, + { + "epoch": 3.4147368421052633, + "grad_norm": 0.67578125, + "learning_rate": 1.0343608793468795e-05, + "loss": 0.7828, + "step": 811 + }, + { + "epoch": 3.4189473684210525, + "grad_norm": 0.5625, + "learning_rate": 1.019576900984599e-05, + "loss": 0.5662, + "step": 812 + }, + { + "epoch": 3.423157894736842, + "grad_norm": 0.69140625, + "learning_rate": 1.0048936599581294e-05, + "loss": 0.6854, + "step": 813 + }, + { + "epoch": 3.4273684210526314, + "grad_norm": 0.5859375, + "learning_rate": 9.903113209758096e-06, + "loss": 0.9926, + "step": 814 + }, + { + "epoch": 3.431578947368421, + "grad_norm": 0.6171875, + "learning_rate": 9.75830047614117e-06, + "loss": 0.9336, + "step": 815 + }, + { + "epoch": 3.4357894736842107, + "grad_norm": 0.56640625, + "learning_rate": 9.614500023158336e-06, + "loss": 0.5855, + "step": 816 + }, + { + "epoch": 3.44, + "grad_norm": 0.609375, + "learning_rate": 9.471713463882193e-06, + "loss": 0.6517, + "step": 817 + }, + { + "epoch": 3.4442105263157896, + "grad_norm": 0.5390625, + "learning_rate": 9.32994240001206e-06, + "loss": 0.7409, + "step": 818 + }, + { + "epoch": 3.448421052631579, + "grad_norm": 0.494140625, + "learning_rate": 9.189188421855976e-06, + "loss": 0.9442, + "step": 819 + }, + { + "epoch": 3.4526315789473685, + "grad_norm": 0.90234375, + "learning_rate": 9.049453108312966e-06, + "loss": 0.9277, + "step": 820 + }, + { + "epoch": 3.4568421052631577, + "grad_norm": 0.462890625, + "learning_rate": 8.91073802685517e-06, + "loss": 0.8213, + "step": 821 + }, + { + "epoch": 3.4610526315789474, + "grad_norm": 0.462890625, + "learning_rate": 8.773044733510338e-06, + "loss": 0.9823, + "step": 822 + }, + { + "epoch": 3.4652631578947366, + "grad_norm": 0.373046875, + "learning_rate": 8.63637477284439e-06, + "loss": 0.7546, + "step": 823 + }, + { + "epoch": 3.4694736842105263, + "grad_norm": 0.51953125, + "learning_rate": 8.50072967794413e-06, + "loss": 0.9332, + "step": 824 + }, + { + "epoch": 3.473684210526316, + "grad_norm": 0.4765625, + "learning_rate": 8.366110970399922e-06, + "loss": 0.8852, + "step": 825 + }, + { + "epoch": 3.477894736842105, + "grad_norm": 0.4921875, + "learning_rate": 8.232520160288704e-06, + "loss": 1.1599, + "step": 826 + }, + { + "epoch": 3.482105263157895, + "grad_norm": 0.72265625, + "learning_rate": 8.099958746157067e-06, + "loss": 0.8591, + "step": 827 + }, + { + "epoch": 3.486315789473684, + "grad_norm": 0.341796875, + "learning_rate": 7.96842821500442e-06, + "loss": 0.8354, + "step": 828 + }, + { + "epoch": 3.4905263157894737, + "grad_norm": 0.70703125, + "learning_rate": 7.837930042266262e-06, + "loss": 1.1108, + "step": 829 + }, + { + "epoch": 3.4947368421052634, + "grad_norm": 0.75390625, + "learning_rate": 7.708465691797717e-06, + "loss": 0.8228, + "step": 830 + }, + { + "epoch": 3.4989473684210526, + "grad_norm": 0.3359375, + "learning_rate": 7.580036615857056e-06, + "loss": 0.6261, + "step": 831 + }, + { + "epoch": 3.5031578947368422, + "grad_norm": 0.52734375, + "learning_rate": 7.452644255089425e-06, + "loss": 0.8411, + "step": 832 + }, + { + "epoch": 3.5073684210526315, + "grad_norm": 0.52734375, + "learning_rate": 7.326290038510686e-06, + "loss": 0.8299, + "step": 833 + }, + { + "epoch": 3.511578947368421, + "grad_norm": 0.4609375, + "learning_rate": 7.20097538349136e-06, + "loss": 0.8227, + "step": 834 + }, + { + "epoch": 3.515789473684211, + "grad_norm": 0.5625, + "learning_rate": 7.076701695740806e-06, + "loss": 0.8436, + "step": 835 + }, + { + "epoch": 3.52, + "grad_norm": 0.400390625, + "learning_rate": 6.953470369291348e-06, + "loss": 0.9071, + "step": 836 + }, + { + "epoch": 3.5242105263157892, + "grad_norm": 0.58984375, + "learning_rate": 6.8312827864826954e-06, + "loss": 0.7809, + "step": 837 + }, + { + "epoch": 3.528421052631579, + "grad_norm": 0.58203125, + "learning_rate": 6.710140317946423e-06, + "loss": 0.817, + "step": 838 + }, + { + "epoch": 3.5326315789473686, + "grad_norm": 0.984375, + "learning_rate": 6.59004432259065e-06, + "loss": 1.0496, + "step": 839 + }, + { + "epoch": 3.536842105263158, + "grad_norm": 0.99609375, + "learning_rate": 6.470996147584685e-06, + "loss": 1.0043, + "step": 840 + }, + { + "epoch": 3.536842105263158, + "eval_loss": 1.2289316654205322, + "eval_runtime": 25.9817, + "eval_samples_per_second": 3.849, + "eval_steps_per_second": 1.924, + "step": 840 + }, + { + "epoch": 3.5410526315789475, + "grad_norm": 0.40625, + "learning_rate": 6.352997128344007e-06, + "loss": 0.9158, + "step": 841 + }, + { + "epoch": 3.5452631578947367, + "grad_norm": 0.54296875, + "learning_rate": 6.236048588515242e-06, + "loss": 0.6722, + "step": 842 + }, + { + "epoch": 3.5494736842105263, + "grad_norm": 0.65234375, + "learning_rate": 6.1201518399613635e-06, + "loss": 0.8325, + "step": 843 + }, + { + "epoch": 3.553684210526316, + "grad_norm": 0.71484375, + "learning_rate": 6.0053081827469045e-06, + "loss": 0.8378, + "step": 844 + }, + { + "epoch": 3.557894736842105, + "grad_norm": 0.47265625, + "learning_rate": 5.891518905123428e-06, + "loss": 0.7611, + "step": 845 + }, + { + "epoch": 3.562105263157895, + "grad_norm": 0.5, + "learning_rate": 5.778785283515053e-06, + "loss": 1.0624, + "step": 846 + }, + { + "epoch": 3.566315789473684, + "grad_norm": 0.703125, + "learning_rate": 5.667108582504143e-06, + "loss": 1.2832, + "step": 847 + }, + { + "epoch": 3.5705263157894738, + "grad_norm": 0.53515625, + "learning_rate": 5.556490054817132e-06, + "loss": 0.5906, + "step": 848 + }, + { + "epoch": 3.5747368421052634, + "grad_norm": 0.41796875, + "learning_rate": 5.446930941310447e-06, + "loss": 0.8559, + "step": 849 + }, + { + "epoch": 3.5789473684210527, + "grad_norm": 0.578125, + "learning_rate": 5.338432470956589e-06, + "loss": 0.9846, + "step": 850 + }, + { + "epoch": 3.583157894736842, + "grad_norm": 1.1328125, + "learning_rate": 5.230995860830412e-06, + "loss": 0.6747, + "step": 851 + }, + { + "epoch": 3.5873684210526315, + "grad_norm": 0.6953125, + "learning_rate": 5.1246223160953845e-06, + "loss": 0.9996, + "step": 852 + }, + { + "epoch": 3.591578947368421, + "grad_norm": 0.484375, + "learning_rate": 5.019313029990092e-06, + "loss": 0.8196, + "step": 853 + }, + { + "epoch": 3.5957894736842104, + "grad_norm": 0.69140625, + "learning_rate": 4.91506918381488e-06, + "loss": 1.1009, + "step": 854 + }, + { + "epoch": 3.6, + "grad_norm": 0.58984375, + "learning_rate": 4.811891946918612e-06, + "loss": 0.7668, + "step": 855 + }, + { + "epoch": 3.6042105263157893, + "grad_norm": 0.6328125, + "learning_rate": 4.7097824766854756e-06, + "loss": 0.9518, + "step": 856 + }, + { + "epoch": 3.608421052631579, + "grad_norm": 0.33984375, + "learning_rate": 4.608741918522097e-06, + "loss": 0.8728, + "step": 857 + }, + { + "epoch": 3.6126315789473686, + "grad_norm": 0.77734375, + "learning_rate": 4.508771405844636e-06, + "loss": 0.4115, + "step": 858 + }, + { + "epoch": 3.616842105263158, + "grad_norm": 0.8203125, + "learning_rate": 4.4098720600660785e-06, + "loss": 0.6579, + "step": 859 + }, + { + "epoch": 3.6210526315789475, + "grad_norm": 0.37890625, + "learning_rate": 4.312044990583675e-06, + "loss": 0.599, + "step": 860 + }, + { + "epoch": 3.6252631578947367, + "grad_norm": 0.4765625, + "learning_rate": 4.215291294766455e-06, + "loss": 0.8502, + "step": 861 + }, + { + "epoch": 3.6294736842105264, + "grad_norm": 0.703125, + "learning_rate": 4.119612057942978e-06, + "loss": 0.7009, + "step": 862 + }, + { + "epoch": 3.6336842105263156, + "grad_norm": 0.76953125, + "learning_rate": 4.025008353389104e-06, + "loss": 0.9114, + "step": 863 + }, + { + "epoch": 3.6378947368421053, + "grad_norm": 0.62890625, + "learning_rate": 3.931481242315993e-06, + "loss": 0.7777, + "step": 864 + }, + { + "epoch": 3.6421052631578945, + "grad_norm": 0.76171875, + "learning_rate": 3.8390317738581595e-06, + "loss": 0.9164, + "step": 865 + }, + { + "epoch": 3.646315789473684, + "grad_norm": 0.875, + "learning_rate": 3.747660985061785e-06, + "loss": 0.6349, + "step": 866 + }, + { + "epoch": 3.650526315789474, + "grad_norm": 0.71484375, + "learning_rate": 3.6573699008729644e-06, + "loss": 0.7808, + "step": 867 + }, + { + "epoch": 3.654736842105263, + "grad_norm": 0.439453125, + "learning_rate": 3.568159534126314e-06, + "loss": 1.1544, + "step": 868 + }, + { + "epoch": 3.6589473684210527, + "grad_norm": 0.9296875, + "learning_rate": 3.480030885533536e-06, + "loss": 1.1083, + "step": 869 + }, + { + "epoch": 3.663157894736842, + "grad_norm": 0.55078125, + "learning_rate": 3.3929849436722728e-06, + "loss": 1.0263, + "step": 870 + }, + { + "epoch": 3.6673684210526316, + "grad_norm": 0.609375, + "learning_rate": 3.3070226849749366e-06, + "loss": 0.7433, + "step": 871 + }, + { + "epoch": 3.6715789473684213, + "grad_norm": 0.55078125, + "learning_rate": 3.2221450737178083e-06, + "loss": 1.0724, + "step": 872 + }, + { + "epoch": 3.6757894736842105, + "grad_norm": 0.63671875, + "learning_rate": 3.1383530620101687e-06, + "loss": 0.4376, + "step": 873 + }, + { + "epoch": 3.68, + "grad_norm": 1.1015625, + "learning_rate": 3.0556475897837166e-06, + "loss": 0.7161, + "step": 874 + }, + { + "epoch": 3.6842105263157894, + "grad_norm": 0.99609375, + "learning_rate": 2.974029584781912e-06, + "loss": 0.6898, + "step": 875 + }, + { + "epoch": 3.688421052631579, + "grad_norm": 0.51171875, + "learning_rate": 2.8934999625496282e-06, + "loss": 0.6731, + "step": 876 + }, + { + "epoch": 3.6926315789473683, + "grad_norm": 0.56640625, + "learning_rate": 2.8140596264228936e-06, + "loss": 1.0746, + "step": 877 + }, + { + "epoch": 3.696842105263158, + "grad_norm": 0.578125, + "learning_rate": 2.735709467518699e-06, + "loss": 1.1134, + "step": 878 + }, + { + "epoch": 3.701052631578947, + "grad_norm": 0.58203125, + "learning_rate": 2.6584503647250847e-06, + "loss": 1.03, + "step": 879 + }, + { + "epoch": 3.705263157894737, + "grad_norm": 0.416015625, + "learning_rate": 2.5822831846912033e-06, + "loss": 1.0116, + "step": 880 + }, + { + "epoch": 3.7094736842105265, + "grad_norm": 0.474609375, + "learning_rate": 2.5072087818176382e-06, + "loss": 1.3308, + "step": 881 + }, + { + "epoch": 3.7136842105263157, + "grad_norm": 0.70703125, + "learning_rate": 2.4332279982468453e-06, + "loss": 0.8459, + "step": 882 + }, + { + "epoch": 3.7178947368421054, + "grad_norm": 0.6875, + "learning_rate": 2.360341663853627e-06, + "loss": 1.0375, + "step": 883 + }, + { + "epoch": 3.7221052631578946, + "grad_norm": 0.453125, + "learning_rate": 2.2885505962359054e-06, + "loss": 0.8915, + "step": 884 + }, + { + "epoch": 3.7263157894736842, + "grad_norm": 0.8671875, + "learning_rate": 2.2178556007054872e-06, + "loss": 0.7733, + "step": 885 + }, + { + "epoch": 3.730526315789474, + "grad_norm": 0.734375, + "learning_rate": 2.1482574702790803e-06, + "loss": 0.8054, + "step": 886 + }, + { + "epoch": 3.734736842105263, + "grad_norm": 0.95703125, + "learning_rate": 2.079756985669379e-06, + "loss": 0.6983, + "step": 887 + }, + { + "epoch": 3.738947368421053, + "grad_norm": 0.71484375, + "learning_rate": 2.0123549152762823e-06, + "loss": 1.1439, + "step": 888 + }, + { + "epoch": 3.743157894736842, + "grad_norm": 0.49609375, + "learning_rate": 1.9460520151783233e-06, + "loss": 1.0331, + "step": 889 + }, + { + "epoch": 3.7473684210526317, + "grad_norm": 0.44140625, + "learning_rate": 1.8808490291241432e-06, + "loss": 0.6042, + "step": 890 + }, + { + "epoch": 3.751578947368421, + "grad_norm": 0.57421875, + "learning_rate": 1.8167466885241625e-06, + "loss": 0.8279, + "step": 891 + }, + { + "epoch": 3.7557894736842106, + "grad_norm": 0.5, + "learning_rate": 1.7537457124423895e-06, + "loss": 0.6845, + "step": 892 + }, + { + "epoch": 3.76, + "grad_norm": 0.5078125, + "learning_rate": 1.6918468075883488e-06, + "loss": 0.9477, + "step": 893 + }, + { + "epoch": 3.7642105263157895, + "grad_norm": 0.57421875, + "learning_rate": 1.631050668309131e-06, + "loss": 0.9711, + "step": 894 + }, + { + "epoch": 3.768421052631579, + "grad_norm": 0.46875, + "learning_rate": 1.5713579765816555e-06, + "loss": 0.6749, + "step": 895 + }, + { + "epoch": 3.7726315789473683, + "grad_norm": 0.59375, + "learning_rate": 1.5127694020049432e-06, + "loss": 0.8045, + "step": 896 + }, + { + "epoch": 3.776842105263158, + "grad_norm": 0.50390625, + "learning_rate": 1.4552856017926886e-06, + "loss": 0.8082, + "step": 897 + }, + { + "epoch": 3.781052631578947, + "grad_norm": 0.671875, + "learning_rate": 1.3989072207658328e-06, + "loss": 0.8721, + "step": 898 + }, + { + "epoch": 3.785263157894737, + "grad_norm": 0.7109375, + "learning_rate": 1.3436348913453578e-06, + "loss": 0.9514, + "step": 899 + }, + { + "epoch": 3.7894736842105265, + "grad_norm": 0.79296875, + "learning_rate": 1.2894692335451375e-06, + "loss": 1.128, + "step": 900 + }, + { + "epoch": 3.7894736842105265, + "eval_loss": 1.2294663190841675, + "eval_runtime": 25.9303, + "eval_samples_per_second": 3.856, + "eval_steps_per_second": 1.928, + "step": 900 + }, + { + "epoch": 3.7936842105263158, + "grad_norm": 0.474609375, + "learning_rate": 1.2364108549650976e-06, + "loss": 0.8639, + "step": 901 + }, + { + "epoch": 3.797894736842105, + "grad_norm": 0.51171875, + "learning_rate": 1.1844603507842668e-06, + "loss": 0.8217, + "step": 902 + }, + { + "epoch": 3.8021052631578947, + "grad_norm": 0.376953125, + "learning_rate": 1.1336183037541914e-06, + "loss": 1.0223, + "step": 903 + }, + { + "epoch": 3.8063157894736843, + "grad_norm": 0.61328125, + "learning_rate": 1.083885284192354e-06, + "loss": 0.8096, + "step": 904 + }, + { + "epoch": 3.8105263157894735, + "grad_norm": 0.53515625, + "learning_rate": 1.0352618499758216e-06, + "loss": 0.9039, + "step": 905 + }, + { + "epoch": 3.814736842105263, + "grad_norm": 0.345703125, + "learning_rate": 9.877485465349058e-07, + "loss": 0.9409, + "step": 906 + }, + { + "epoch": 3.8189473684210524, + "grad_norm": 2.015625, + "learning_rate": 9.413459068471353e-07, + "loss": 1.2651, + "step": 907 + }, + { + "epoch": 3.823157894736842, + "grad_norm": 0.5546875, + "learning_rate": 8.960544514312275e-07, + "loss": 0.8617, + "step": 908 + }, + { + "epoch": 3.8273684210526318, + "grad_norm": 0.69140625, + "learning_rate": 8.518746883412587e-07, + "loss": 0.9737, + "step": 909 + }, + { + "epoch": 3.831578947368421, + "grad_norm": 0.609375, + "learning_rate": 8.088071131609587e-07, + "loss": 0.8379, + "step": 910 + }, + { + "epoch": 3.8357894736842106, + "grad_norm": 0.67578125, + "learning_rate": 7.66852208998159e-07, + "loss": 0.9968, + "step": 911 + }, + { + "epoch": 3.84, + "grad_norm": 0.76953125, + "learning_rate": 7.26010446479397e-07, + "loss": 0.9101, + "step": 912 + }, + { + "epoch": 3.8442105263157895, + "grad_norm": 0.482421875, + "learning_rate": 6.862822837445881e-07, + "loss": 1.0709, + "step": 913 + }, + { + "epoch": 3.848421052631579, + "grad_norm": 0.57421875, + "learning_rate": 6.476681664419171e-07, + "loss": 0.7148, + "step": 914 + }, + { + "epoch": 3.8526315789473684, + "grad_norm": 0.470703125, + "learning_rate": 6.101685277228319e-07, + "loss": 0.9929, + "step": 915 + }, + { + "epoch": 3.8568421052631576, + "grad_norm": 0.53515625, + "learning_rate": 5.737837882371922e-07, + "loss": 0.6993, + "step": 916 + }, + { + "epoch": 3.8610526315789473, + "grad_norm": 0.5390625, + "learning_rate": 5.3851435612855e-07, + "loss": 1.0105, + "step": 917 + }, + { + "epoch": 3.865263157894737, + "grad_norm": 0.45703125, + "learning_rate": 5.043606270295654e-07, + "loss": 1.1952, + "step": 918 + }, + { + "epoch": 3.869473684210526, + "grad_norm": 0.50390625, + "learning_rate": 4.7132298405754285e-07, + "loss": 0.9961, + "step": 919 + }, + { + "epoch": 3.873684210526316, + "grad_norm": 0.43359375, + "learning_rate": 4.3940179781019055e-07, + "loss": 1.1113, + "step": 920 + }, + { + "epoch": 3.877894736842105, + "grad_norm": 0.51953125, + "learning_rate": 4.0859742636142337e-07, + "loss": 1.0482, + "step": 921 + }, + { + "epoch": 3.8821052631578947, + "grad_norm": 0.30078125, + "learning_rate": 3.789102152573665e-07, + "loss": 0.693, + "step": 922 + }, + { + "epoch": 3.8863157894736844, + "grad_norm": 0.455078125, + "learning_rate": 3.503404975124469e-07, + "loss": 0.9989, + "step": 923 + }, + { + "epoch": 3.8905263157894736, + "grad_norm": 0.3828125, + "learning_rate": 3.228885936056858e-07, + "loss": 0.8858, + "step": 924 + }, + { + "epoch": 3.8947368421052633, + "grad_norm": 2.03125, + "learning_rate": 2.9655481147712326e-07, + "loss": 0.9219, + "step": 925 + }, + { + "epoch": 3.8989473684210525, + "grad_norm": 0.52734375, + "learning_rate": 2.713394465242991e-07, + "loss": 0.9195, + "step": 926 + }, + { + "epoch": 3.903157894736842, + "grad_norm": 0.443359375, + "learning_rate": 2.472427815989886e-07, + "loss": 0.9765, + "step": 927 + }, + { + "epoch": 3.907368421052632, + "grad_norm": 0.37890625, + "learning_rate": 2.242650870040497e-07, + "loss": 0.9362, + "step": 928 + }, + { + "epoch": 3.911578947368421, + "grad_norm": 0.765625, + "learning_rate": 2.024066204903252e-07, + "loss": 0.5604, + "step": 929 + }, + { + "epoch": 3.9157894736842103, + "grad_norm": 1.15625, + "learning_rate": 1.8166762725381203e-07, + "loss": 1.0528, + "step": 930 + }, + { + "epoch": 3.92, + "grad_norm": 0.455078125, + "learning_rate": 1.6204833993285206e-07, + "loss": 0.9142, + "step": 931 + }, + { + "epoch": 3.9242105263157896, + "grad_norm": 0.57421875, + "learning_rate": 1.4354897860558992e-07, + "loss": 1.2521, + "step": 932 + }, + { + "epoch": 3.928421052631579, + "grad_norm": 0.57421875, + "learning_rate": 1.2616975078746373e-07, + "loss": 0.9689, + "step": 933 + }, + { + "epoch": 3.9326315789473685, + "grad_norm": 0.42578125, + "learning_rate": 1.0991085142886271e-07, + "loss": 0.7758, + "step": 934 + }, + { + "epoch": 3.9368421052631577, + "grad_norm": 0.8203125, + "learning_rate": 9.477246291297314e-08, + "loss": 0.9825, + "step": 935 + }, + { + "epoch": 3.9410526315789474, + "grad_norm": 0.5, + "learning_rate": 8.075475505373575e-08, + "loss": 0.9046, + "step": 936 + }, + { + "epoch": 3.945263157894737, + "grad_norm": 0.4765625, + "learning_rate": 6.785788509392488e-08, + "loss": 1.0151, + "step": 937 + }, + { + "epoch": 3.9494736842105262, + "grad_norm": 0.47265625, + "learning_rate": 5.608199770334999e-08, + "loss": 0.873, + "step": 938 + }, + { + "epoch": 3.953684210526316, + "grad_norm": 0.578125, + "learning_rate": 4.542722497730134e-08, + "loss": 0.8194, + "step": 939 + }, + { + "epoch": 3.957894736842105, + "grad_norm": 0.703125, + "learning_rate": 3.5893686435029e-08, + "loss": 0.9057, + "step": 940 + }, + { + "epoch": 3.962105263157895, + "grad_norm": 0.62109375, + "learning_rate": 2.7481489018410523e-08, + "loss": 1.1382, + "step": 941 + }, + { + "epoch": 3.9663157894736845, + "grad_norm": 0.380859375, + "learning_rate": 2.019072709074088e-08, + "loss": 1.0322, + "step": 942 + }, + { + "epoch": 3.9705263157894737, + "grad_norm": 0.5078125, + "learning_rate": 1.4021482435688794e-08, + "loss": 0.9812, + "step": 943 + }, + { + "epoch": 3.974736842105263, + "grad_norm": 0.49609375, + "learning_rate": 8.973824256364171e-09, + "loss": 0.7316, + "step": 944 + }, + { + "epoch": 3.9789473684210526, + "grad_norm": 0.546875, + "learning_rate": 5.047809174574258e-09, + "loss": 0.9063, + "step": 945 + }, + { + "epoch": 3.9831578947368422, + "grad_norm": 0.421875, + "learning_rate": 2.2434812301352913e-09, + "loss": 1.0714, + "step": 946 + }, + { + "epoch": 3.9873684210526315, + "grad_norm": 0.474609375, + "learning_rate": 5.60871880417313e-10, + "loss": 0.9579, + "step": 947 + }, + { + "epoch": 3.991578947368421, + "grad_norm": 0.859375, + "learning_rate": 0.0, + "loss": 0.8667, + "step": 948 + } + ], + "logging_steps": 1, + "max_steps": 948, + "num_input_tokens_seen": 0, + "num_train_epochs": 4, + "save_steps": 237, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 1.5790961756995584e+16, + "train_batch_size": 2, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-948/training_args.bin b/checkpoint-948/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..bd7f9976bbc503fbce44b538eec2dade4024c02a --- /dev/null +++ b/checkpoint-948/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee017c711219d88132625f3e50d1a67d1c6236a55d3f5a2ad9f499310d968cef +size 5944 diff --git a/config.json b/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4fd46aded17c1e2a61e180b1bb90c4097c007cd9 --- /dev/null +++ b/config.json @@ -0,0 +1,44 @@ +{ + "_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 1, + "eos_token_id": 2, + "hidden_act": "silu", + "hidden_size": 2048, + "initializer_range": 0.02, + "intermediate_size": 5632, + "max_position_embeddings": 4096, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 22, + "num_key_value_heads": 4, + "pretraining_tp": 1, + "quantization_config": { + "_load_in_4bit": true, + "_load_in_8bit": false, + "bnb_4bit_compute_dtype": "float32", + "bnb_4bit_quant_storage": "bfloat16", + "bnb_4bit_quant_type": "nf4", + "bnb_4bit_use_double_quant": true, + "llm_int8_enable_fp32_cpu_offload": false, + "llm_int8_has_fp16_weight": false, + "llm_int8_skip_modules": null, + "llm_int8_threshold": 6.0, + "load_in_4bit": true, + "load_in_8bit": false, + "quant_method": "bitsandbytes" + }, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 10000.0, + "tie_word_embeddings": false, + "torch_dtype": "float32", + "transformers_version": "4.41.1", + "use_cache": false, + "vocab_size": 32000 +} diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,24 @@ +{ + "bos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": "", + "unk_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.model b/tokenizer.model new file mode 100644 index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899 --- /dev/null +++ b/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 +size 499723 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..0773857a13ba5a27453a0b462624fe76e8e82a86 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,44 @@ +{ + "add_bos_token": true, + "add_eos_token": false, + "add_prefix_space": true, + "added_tokens_decoder": { + "0": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "1": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "2": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "", + "clean_up_tokenization_spaces": false, + "eos_token": "", + "legacy": false, + "model_max_length": 1000000000000000019884624838656, + "pad_token": "", + "padding_side": "right", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "LlamaTokenizer", + "unk_token": "", + "use_default_system_prompt": false, + "use_fast": true +}