Split (1)

train · 844k rows

modelId stringlengths 5 139	author stringlengths 2 42	last_modified timestamp[us, tz=UTC]date 2020-02-15 11:33:14 2025-06-26 18:27:55	downloads int64 0 223M	likes int64 0 11.7k	library_name stringclasses 499 values	tags sequencelengths 1 4.05k	pipeline_tag stringclasses 54 values	createdAt timestamp[us, tz=UTC]date 2022-03-02 23:29:04 2025-06-26 18:27:32	card stringlengths 11 1.01M
nicoboss/Qwen-3-32B-Medical-Reasoning	nicoboss	2025-05-02T12:00:43Z	0	0	transformers	[ "transformers", "safetensors", "qwen3", "text-generation", "medical", "xnet", "qwen", "conversational", "en", "dataset:FreedomIntelligence/medical-o1-reasoning-SFT", "base_model:Qwen/Qwen3-32B", "base_model:finetune:Qwen/Qwen3-32B", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T11:20:56Z	--- library_name: transformers tags: - medical - xnet - qwen license: apache-2.0 datasets: - FreedomIntelligence/medical-o1-reasoning-SFT language: - en base_model: - Qwen/Qwen3-32B pipeline_tag: text-generation --- # Intro This is https://huggingface.co/kingabzpro/Qwen-3-32B-Medical-Reasoning applied to https://huggingface.co/Qwen/Qwen3-32B [Original model card](https://huggingface.co/kingabzpro/Qwen-3-32B-Medical-Reasoning) created by [@kingabzpro](https://huggingface.co/kingabzpro) --- # Original model card from @kingabzpro ## Fine-tuning Qwen3-32B in 4-bit Quantization for Medical Reasoning This project fine-tunes the [`Qwen/Qwen3-32B`](https://huggingface.co/Qwen/Qwen3-32B) model using a medical reasoning dataset (`FreedomIntelligence/medical-o1-reasoning-SFT`) with 4-bit quantization for memory-efficient training. --- ## Setup 1. Install the required libraries: ```bash pip install -U datasets accelerate peft trl bitsandbytes pip install -U transformers pip install huggingface_hub[hf_xet] ``` 2. Authenticate with Hugging Face Hub: Make sure your Hugging Face token is stored in an environment variable: ```bash export HF_TOKEN=your_huggingface_token ``` The notebook will automatically log you in using this token. --- ## How to Run 1. Load the Model and Tokenizer The script downloads the Qwen3-32B model and applies 4-bit quantization with `BitsAndBytesConfig` for efficient memory usage. 2. Prepare the Dataset - The notebook uses `FreedomIntelligence/medical-o1-reasoning-SFT` (first 500 samples). - It formats each example into an instruction-following prompt with step-by-step chain-of-thought reasoning. 3. Fine-tuning - Fine-tuning is set up with PEFT (LoRA / Adapter Tuning style) to modify a small subset of model parameters. - TRL (Transformer Reinforcement Learning) is used to fine-tune efficiently. 4. Push Fine-tuned Model - After training, the fine-tuned model and tokenizer are pushed back to your Hugging Face account. --- >> Here is the training notebook: [Fine_tuning_Qwen-3-32B](https://huggingface.co/kingabzpro/Qwen-3-32B-Medical-Reasoning/blob/main/fine-tuning-qwen-3.ipynb) ## Model Configuration - Base Model: `Qwen/Qwen3-32B` - Quantization: 4-bit (NF4) - Training: PEFT + TRL - Dataset: 2000 examples from medical reasoning dataset --- ## Notes - GPU Required: Make sure you have access to 1X A100s. Get it from RunPod for an hours. Training took only 50 minutes. - Environment: The notebook expects an environment where NVIDIA CUDA drivers are available (`nvidia-smi` check is included). - Memory Efficiency: 4-bit loading greatly reduces memory footprint. --- ## Example Prompt Format ``` Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request. Before answering, think carefully about the question and create a step-by-step chain of thoughts to ensure a logical and accurate response. ### Instruction: You are a medical expert with advanced knowledge in clinical reasoning, diagnostics, and treatment planning. Please answer the following medical question. ### Question: {} ### Response: <think> {} </think> {} ``` --- ## Usage Script (not-tested) ```python from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig from peft import PeftModel import torch # Base model (original model from Meta) base_model_id = "Qwen/Qwen3-32B" # Your fine-tuned LoRA adapter repository lora_adapter_id = "kingabzpro/Qwen-3-32B-Medical-Reasoning" # Load the model in 4-bit bnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_use_double_quant=False, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.bfloat16, ) # Load base model base_model = AutoModelForCausalLM.from_pretrained( base_model_id, device_map="auto", torch_dtype=torch.bfloat16, quantization_config=bnb_config, trust_remote_code=True, ) # Attach the LoRA adapter model = PeftModel.from_pretrained( base_model, lora_adapter_id, device_map="auto", trust_remote_code=True, ) # Load tokenizer tokenizer = AutoTokenizer.from_pretrained(base_model_id, trust_remote_code=True) # Inference example prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request. Before answering, think carefully about the question and create a step-by-step chain of thoughts to ensure a logical and accurate response. ### Instruction: You are a medical expert with advanced knowledge in clinical reasoning, diagnostics, and treatment planning. Please answer the following medical question. ### Question: What is the initial management for a patient presenting with diabetic ketoacidosis (DKA)? ### Response: <think> """ inputs = tokenizer(prompt, return_tensors="pt").to(model.device) outputs = model.generate(**inputs, max_new_tokens=1200) response = tokenizer.decode(outputs[0], skip_special_tokens=True) print(response) ```
kblz/mms-tts-amh-v2	kblz	2025-05-02T12:00:20Z	0	0	transformers	[ "transformers", "safetensors", "vits", "arxiv:1910.09700", "endpoints_compatible", "region:us" ]	null	2025-05-02T11:59:40Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
zelk12/MT1-gemma-3-12B	zelk12	2025-05-02T12:00:16Z	0	0	transformers	[ "transformers", "safetensors", "gemma3", "image-text-to-text", "mergekit", "merge", "conversational", "arxiv:2311.03099", "base_model:IlyaGusev/saiga_gemma3_12b", "base_model:merge:IlyaGusev/saiga_gemma3_12b", "base_model:TheDrummer/Fallen-Gemma3-12B-v1", "base_model:merge:TheDrummer/Fallen-Gemma3-12B-v1", "license:gemma", "text-generation-inference", "endpoints_compatible", "region:us" ]	image-text-to-text	2025-05-02T11:50:24Z	--- base_model: - IlyaGusev/saiga_gemma3_12b - TheDrummer/Fallen-Gemma3-12B-v1 library_name: transformers tags: - mergekit - merge license: gemma pipeline_tag: image-text-to-text --- # merge This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). ## Merge Details ### Merge Method This model was merged using the [DARE TIES](https://arxiv.org/abs/2311.03099) merge method using [IlyaGusev/saiga_gemma3_12b](https://huggingface.co/IlyaGusev/saiga_gemma3_12b) as a base. ### Models Merged The following models were included in the merge: * [TheDrummer/Fallen-Gemma3-12B-v1](https://huggingface.co/TheDrummer/Fallen-Gemma3-12B-v1) ### Configuration The following YAML configuration was used to produce this model: ```yaml models: - model: IlyaGusev/saiga_gemma3_12b #no parameters necessary for base model - model: TheDrummer/Fallen-Gemma3-12B-v1 parameters: density: 0.5 weight: 0.5 merge_method: dare_ties base_model: IlyaGusev/saiga_gemma3_12b parameters: normalize: true dtype: bfloat16 ```
showhandshowhand/task-8-microsoft-Phi-3.5-mini-instruct	showhandshowhand	2025-05-02T11:59:36Z	322	0	peft	[ "peft", "safetensors", "arxiv:1910.09700", "base_model:microsoft/Phi-3.5-mini-instruct", "base_model:adapter:microsoft/Phi-3.5-mini-instruct", "region:us" ]	null	2025-05-01T04:41:48Z	--- base_model: microsoft/Phi-3.5-mini-instruct library_name: peft --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed] ### Framework versions - PEFT 0.15.1
seekerdeep/task-8-microsoft-Phi-3.5-mini-instruct	seekerdeep	2025-05-02T11:53:22Z	358	0	peft	[ "peft", "safetensors", "arxiv:1910.09700", "base_model:microsoft/Phi-3.5-mini-instruct", "base_model:adapter:microsoft/Phi-3.5-mini-instruct", "region:us" ]	null	2025-04-13T05:07:28Z	--- base_model: microsoft/Phi-3.5-mini-instruct library_name: peft --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed] ### Framework versions - PEFT 0.15.1
phospho-app/Starkosaure-Follow_The_Line_V0.0-f789hxnp37	phospho-app	2025-05-02T11:53:09Z	0	0	null	[ "phosphobot", "gr00t", "region:us" ]	null	2025-05-02T11:51:57Z	--- tags: - phosphobot - gr00t task_categories: - robotics --- # gr00t Model - phospho Training Pipeline ## Error Traceback We faced an issue while training your model. ``` Traceback (most recent call last): File "/root/src/helper.py", line 224, in predict raise RuntimeError(error_msg) RuntimeError: Training process failed with exit code 1: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/workspace/gr00t/data/dataset.py", line 644, in get_video trajectory_index = self.get_trajectory_index(trajectory_id) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/workspace/gr00t/data/dataset.py", line 557, in get_trajectory_index raise ValueError( ValueError: Error finding trajectory index for 40, found trajectory_indices=array([84, 85, 86, 87]) 0%\| \| 0/400 [00:01<?, ?it/s] During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/root/src/helper.py", line 226, in predict raise RuntimeError(e) RuntimeError: Training process failed with exit code 1: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/workspace/gr00t/data/dataset.py", line 644, in get_video trajectory_index = self.get_trajectory_index(trajectory_id) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/workspace/gr00t/data/dataset.py", line 557, in get_trajectory_index raise ValueError( ValueError: Error finding trajectory index for 40, found trajectory_indices=array([84, 85, 86, 87]) 0%\| \| 0/400 [00:01<?, ?it/s] ``` ## Training parameters: - Dataset: [Starkosaure/Follow_The_Line_V0.0](https://huggingface.co/datasets/Starkosaure/Follow_The_Line_V0.0) - Wandb run URL: None - Epochs: 10 - Batch size: 64 - Training steps: 406 📖 Get Started: [docs.phospho.ai](https://docs.phospho.ai?utm_source=replicate_groot_training_pipeline) 🤖 Get your robot: [robots.phospho.ai](https://robots.phospho.ai?utm_source=replicate_groot_training_pipeline)
MinaMila/phi3_unlearned_LoRa_ACSEmployment_2_ep6_22	MinaMila	2025-05-02T11:52:33Z	0	0	transformers	[ "transformers", "safetensors", "arxiv:1910.09700", "endpoints_compatible", "region:us" ]	null	2025-05-02T11:52:30Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
luckycanucky/hnd_model_16b2	luckycanucky	2025-05-02T11:51:34Z	0	0	transformers	[ "transformers", "pytorch", "llama", "text-generation", "text-generation-inference", "unsloth", "trl", "sft", "conversational", "en", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T11:48:36Z	--- base_model: unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - llama - trl - sft license: apache-2.0 language: - en --- # Uploaded model - Developed by: luckycanucky - License: apache-2.0 - Finetuned from model : unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
yusuke111/myBit-Llama2-jp-127M-2B4TLike-2024	yusuke111	2025-05-02T11:50:25Z	4	0	transformers	[ "transformers", "safetensors", "bit_llama", "text-generation", "generated_from_trainer", "custom_code", "autotrain_compatible", "region:us" ]	text-generation	2025-04-30T15:05:17Z	--- library_name: transformers tags: - generated_from_trainer model-index: - name: myBit-Llama2-jp-127M-2B4TLike results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # myBit-Llama2-jp-127M-2B4TLike This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 2.8431 ## 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: 0.0024 - train_batch_size: 12 - eval_batch_size: 12 - seed: 42 - gradient_accumulation_steps: 8 - total_train_batch_size: 96 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 750 - num_epochs: 1 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 4.9846 \| 0.0587 \| 500 \| 5.1982 \| \| 3.7747 \| 0.1175 \| 1000 \| 4.4941 \| \| 3.5109 \| 0.1762 \| 1500 \| 4.0737 \| \| 3.3568 \| 0.2350 \| 2000 \| 3.8909 \| \| 3.276 \| 0.2937 \| 2500 \| 3.7147 \| \| 3.2203 \| 0.3525 \| 3000 \| 3.5468 \| \| 3.1626 \| 0.4112 \| 3500 \| 3.4098 \| \| 3.1272 \| 0.4700 \| 4000 \| 3.3188 \| \| 3.0925 \| 0.5287 \| 4500 \| 3.2339 \| \| 3.0693 \| 0.5874 \| 5000 \| 3.1539 \| \| 3.0412 \| 0.6462 \| 5500 \| 3.0721 \| \| 2.9981 \| 0.7049 \| 6000 \| 3.0009 \| \| 2.9881 \| 0.7637 \| 6500 \| 2.9514 \| \| 2.9871 \| 0.8224 \| 7000 \| 2.9162 \| \| 2.9796 \| 0.8812 \| 7500 \| 2.8879 \| \| 2.9914 \| 0.9399 \| 8000 \| 2.8849 \| \| 2.9649 \| 0.9987 \| 8500 \| 2.8431 \| ### Framework versions - Transformers 4.47.1 - Pytorch 2.6.0+cu124 - Datasets 3.5.1 - Tokenizers 0.21.1
zelk12/MT-gemma-3-12B	zelk12	2025-05-02T11:47:10Z	0	0	transformers	[ "transformers", "safetensors", "gemma3", "image-text-to-text", "mergekit", "merge", "conversational", "arxiv:2311.03099", "base_model:IlyaGusev/saiga_gemma3_12b", "base_model:merge:IlyaGusev/saiga_gemma3_12b", "base_model:soob3123/amoral-gemma3-12B-v2", "base_model:merge:soob3123/amoral-gemma3-12B-v2", "license:gemma", "text-generation-inference", "endpoints_compatible", "region:us" ]	image-text-to-text	2025-05-02T11:41:19Z	--- base_model: - IlyaGusev/saiga_gemma3_12b - soob3123/amoral-gemma3-12B-v2 library_name: transformers tags: - mergekit - merge license: gemma pipeline_tag: image-text-to-text --- # merge This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). ## Merge Details ### Merge Method This model was merged using the [DARE TIES](https://arxiv.org/abs/2311.03099) merge method using [soob3123/amoral-gemma3-12B-v2](https://huggingface.co/soob3123/amoral-gemma3-12B-v2) as a base. ### Models Merged The following models were included in the merge: * [IlyaGusev/saiga_gemma3_12b](https://huggingface.co/IlyaGusev/saiga_gemma3_12b) ### Configuration The following YAML configuration was used to produce this model: ```yaml models: - model: soob3123/amoral-gemma3-12B-v2 #no parameters necessary for base model - model: IlyaGusev/saiga_gemma3_12b parameters: density: 0.5 weight: 0.5 merge_method: dare_ties base_model: soob3123/amoral-gemma3-12B-v2 parameters: normalize: true dtype: bfloat16 ```
AdoCleanCode/soft_real_imagenet_v7	AdoCleanCode	2025-05-02T11:45:50Z	0	0	transformers	[ "transformers", "tensorboard", "safetensors", "gpt2", "text-generation", "generated_from_trainer", "base_model:openai-community/gpt2", "base_model:finetune:openai-community/gpt2", "license:mit", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T10:47:18Z	--- library_name: transformers license: mit base_model: gpt2 tags: - generated_from_trainer model-index: - name: soft_real_imagenet_v7 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # soft_real_imagenet_v7 This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.9272 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 16 - seed: 42 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1000 - num_epochs: 5 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:-----:\|:----:\|:---------------:\| \| 1.3667 \| 1.0 \| 1576 \| 1.1707 \| \| 1.1367 \| 2.0 \| 3152 \| 1.0222 \| \| 1.0318 \| 3.0 \| 4728 \| 0.9648 \| \| 0.9827 \| 4.0 \| 6304 \| 0.9362 \| \| 0.9406 \| 5.0 \| 7880 \| 0.9272 \| ### Framework versions - Transformers 4.46.3 - Pytorch 2.4.1+cu121 - Datasets 2.19.1 - Tokenizers 0.20.3
Clip-Full-18-Valerie-Kay-Video-Origina/Clip-Full-18-Valerie-Kay-Video-Original-Link-Tikto	Clip-Full-18-Valerie-Kay-Video-Origina	2025-05-02T11:44:33Z	0	0	null	[ "region:us" ]	null	2025-05-02T11:42:43Z	Watch 🟢 ➤ ➤ ➤ <a href="https://sulkurl.com/pg0"> 🌐 Click Here To link (Full Viral Video Link) 🔴 ➤►DOWNLOAD👉👉🟢 ➤ Watch 🟢 ➤ ➤ ➤ <a href="https://sulkurl.com/pg0"> 🌐 Click Here To link (Full Viral Video Link) 🔴 ➤►DOWNLOAD👉👉🟢 ➤
naveennagar0909/lora-coke-can-flux-1dev-run1	naveennagar0909	2025-05-02T11:44:05Z	0	0	diffusers	[ "diffusers", "text-to-image", "diffusers-training", "lora", "flux", "flux-diffusers", "template:sd-lora", "base_model:black-forest-labs/FLUX.1-schnell", "base_model:adapter:black-forest-labs/FLUX.1-schnell", "license:other", "region:us" ]	text-to-image	2025-05-02T11:18:01Z	--- base_model: black-forest-labs/FLUX.1-schnell library_name: diffusers license: other instance_prompt: a photo of sks coke can widget: - text: A photo of sks coke can on a mountain output: url: image_0.png tags: - text-to-image - diffusers-training - diffusers - lora - flux - flux-diffusers - template:sd-lora --- <!-- This model card has been generated automatically according to the information the training script had access to. You should probably proofread and complete it, then remove this comment. --> # Flux DreamBooth LoRA - naveennagar0909/lora-coke-can-flux-1dev-run1 <Gallery /> ## Model description These are naveennagar0909/lora-coke-can-flux-1dev-run1 DreamBooth LoRA weights for black-forest-labs/FLUX.1-schnell. The weights were trained using [DreamBooth](https://dreambooth.github.io/) with the [Flux diffusers trainer](https://github.com/huggingface/diffusers/blob/main/examples/dreambooth/README_flux.md). Was LoRA for the text encoder enabled? False. ## Trigger words You should use `a photo of sks coke can` to trigger the image generation. ## Download model [Download the *.safetensors LoRA](naveennagar0909/lora-coke-can-flux-1dev-run1/tree/main) in the Files & versions tab. ## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers) ```py from diffusers import AutoPipelineForText2Image import torch pipeline = AutoPipelineForText2Image.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16).to('cuda') pipeline.load_lora_weights('naveennagar0909/lora-coke-can-flux-1dev-run1', weight_name='pytorch_lora_weights.safetensors') image = pipeline('A photo of sks coke can on a mountain').images[0] ``` For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters) ## License Please adhere to the licensing terms as described [here](https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md). ## Intended uses & limitations #### How to use ```python # TODO: add an example code snippet for running this diffusion pipeline ``` #### Limitations and bias [TODO: provide examples of latent issues and potential remediations] ## Training details [TODO: describe the data used to train the model]
lisabdunlap/Llama-3.1-8B-Instruct-unsloth-bnb-4bit-r32-e20-lr0.0002-markdown_format_small-new	lisabdunlap	2025-05-02T11:38:49Z	0	0	transformers	[ "transformers", "safetensors", "llama", "text-generation", "text-generation-inference", "unsloth", "trl", "conversational", "en", "base_model:unsloth/Llama-3.1-8B-Instruct-unsloth-bnb-4bit", "base_model:finetune:unsloth/Llama-3.1-8B-Instruct-unsloth-bnb-4bit", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T11:36:43Z	--- base_model: unsloth/Llama-3.1-8B-Instruct-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - llama - trl license: apache-2.0 language: - en --- # Uploaded model - Developed by: lisabdunlap - License: apache-2.0 - Finetuned from model : unsloth/Llama-3.1-8B-Instruct-unsloth-bnb-4bit This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
mcstaykahan/sdvsdv	mcstaykahan	2025-05-02T11:36:30Z	0	0	null	[ "license:bsd-3-clause-clear", "region:us" ]	null	2025-05-02T11:36:30Z	--- license: bsd-3-clause-clear ---
noobjp/testmodel	noobjp	2025-05-02T11:35:27Z	0	0	transformers	[ "transformers", "pytorch", "qwen3", "text-generation", "text-generation-inference", "unsloth", "trl", "sft", "conversational", "en", "base_model:unsloth/Qwen3-0.6B-unsloth-bnb-4bit", "base_model:finetune:unsloth/Qwen3-0.6B-unsloth-bnb-4bit", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T11:34:56Z	--- base_model: unsloth/Qwen3-0.6B-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - qwen3 - trl - sft license: apache-2.0 language: - en --- # Uploaded model - Developed by: noobjp - License: apache-2.0 - Finetuned from model : unsloth/Qwen3-0.6B-unsloth-bnb-4bit This qwen3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
nis12ram/qwen3-4b-hindiNER-all-exp2-new	nis12ram	2025-05-02T11:33:08Z	0	0	transformers	[ "transformers", "safetensors", "qwen3", "text-generation", "conversational", "arxiv:1910.09700", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T11:30:08Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
taobao-mnn/DeepSeek-Prover-V2-7B-MNN	taobao-mnn	2025-05-02T11:32:58Z	0	0	null	[ "chat", "text-generation", "en", "license:apache-2.0", "region:us" ]	text-generation	2025-05-02T02:43:54Z	--- license: apache-2.0 language: - en pipeline_tag: text-generation tags: - chat --- # DeepSeek-Prover-V2-7B-MNN ## Introduction This model is a 4-bit quantized version of the MNN model exported from DeepSeek-Prover-V2-7B using [llmexport](https://github.com/alibaba/MNN/tree/master/transformers/llm/export). ## Download ```bash # install huggingface pip install huggingface ``` ```bash # shell download huggingface download --model 'taobao-mnn/DeepSeek-Prover-V2-7B-MNN' --local_dir 'path/to/dir' ``` ```python # SDK download from huggingface_hub import snapshot_download model_dir = snapshot_download('taobao-mnn/DeepSeek-Prover-V2-7B-MNN') ``` ```bash # git clone git clone https://www.modelscope.cn/taobao-mnn/DeepSeek-Prover-V2-7B-MNN ``` ## Usage ```bash # clone MNN source git clone https://github.com/alibaba/MNN.git # compile cd MNN mkdir build && cd build cmake .. -DMNN_LOW_MEMORY=true -DMNN_CPU_WEIGHT_DEQUANT_GEMM=true -DMNN_BUILD_LLM=true -DMNN_SUPPORT_TRANSFORMER_FUSE=true make -j # run ./llm_demo /path/to/DeepSeek-Prover-V2-7B-MNN/config.json prompt.txt ``` ## Document [MNN-LLM](https://mnn-docs.readthedocs.io/en/latest/transformers/llm.html#)
mergekit-community/mergekit-dare_ties-bqkjlyo	mergekit-community	2025-05-02T11:32:32Z	0	0	transformers	[ "transformers", "safetensors", "gemma3", "image-text-to-text", "mergekit", "merge", "conversational", "arxiv:2311.03099", "base_model:IlyaGusev/saiga_gemma3_12b", "base_model:finetune:IlyaGusev/saiga_gemma3_12b", "text-generation-inference", "endpoints_compatible", "region:us" ]	image-text-to-text	2025-05-02T11:28:00Z	--- base_model: - IlyaGusev/saiga_gemma3_12b library_name: transformers tags: - mergekit - merge --- # merge This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). ## Merge Details ### Merge Method This model was merged using the [DARE TIES](https://arxiv.org/abs/2311.03099) merge method using [IlyaGusev/saiga_gemma3_12b](https://huggingface.co/IlyaGusev/saiga_gemma3_12b) as a base. ### Models Merged The following models were included in the merge: ### Configuration The following YAML configuration was used to produce this model: ```yaml models: - model: IlyaGusev/saiga_gemma3_12b #no parameters necessary for base model - model: IlyaGusev/saiga_gemma3_12b parameters: density: 0.5 weight: 0.5 merge_method: dare_ties base_model: IlyaGusev/saiga_gemma3_12b parameters: normalize: true dtype: bfloat16 ```
Ahresh-53/GAMGAM	Ahresh-53	2025-05-02T11:32:29Z	0	0	null	[ "license:apache-2.0", "region:us" ]	null	2025-05-02T11:32:29Z	--- license: apache-2.0 ---
kavlab/qwen3-4B-text-to-1csql	kavlab	2025-05-02T11:31:55Z	0	0	transformers	[ "transformers", "safetensors", "arxiv:1910.09700", "endpoints_compatible", "region:us" ]	null	2025-05-02T11:29:44Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
alsatesamsol/xcvxcv	alsatesamsol	2025-05-02T11:27:32Z	0	0	null	[ "license:bigscience-openrail-m", "region:us" ]	null	2025-05-02T11:27:32Z	--- license: bigscience-openrail-m ---
aryan7777777/deepseek-finetuned-on-osc-data	aryan7777777	2025-05-02T11:23:18Z	0	0	null	[ "safetensors", "llama", "unsloth", "trl", "sft", "license:mit", "region:us" ]	null	2025-05-02T10:56:26Z	--- license: mit tags: - unsloth - trl - sft ---
AdoCleanCode/real_model_CI10_v4	AdoCleanCode	2025-05-02T11:20:45Z	0	0	transformers	[ "transformers", "tensorboard", "safetensors", "gpt2", "text-generation", "generated_from_trainer", "base_model:openai-community/gpt2", "base_model:finetune:openai-community/gpt2", "license:mit", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T09:35:05Z	--- library_name: transformers license: mit base_model: gpt2 tags: - generated_from_trainer model-index: - name: real_model_CI10_v4 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # real_model_CI10_v4 This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.3957 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 16 - seed: 42 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1000 - num_epochs: 5 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:-----:\|:----:\|:---------------:\| \| 0.6216 \| 1.0 \| 1810 \| 0.4910 \| \| 0.4872 \| 2.0 \| 3620 \| 0.4347 \| \| 0.4516 \| 3.0 \| 5430 \| 0.4114 \| \| 0.4214 \| 4.0 \| 7240 \| 0.4010 \| \| 0.4079 \| 5.0 \| 9050 \| 0.3957 \| ### Framework versions - Transformers 4.46.3 - Pytorch 2.4.1+cu121 - Datasets 2.19.1 - Tokenizers 0.20.3
ASethi04/meta-llama-Llama-3.1-8B-pubmedqa-first-lora-4-0.0001-no-prompt-template	ASethi04	2025-05-02T11:12:44Z	0	0	transformers	[ "transformers", "safetensors", "generated_from_trainer", "trl", "sft", "base_model:meta-llama/Llama-3.1-8B", "base_model:finetune:meta-llama/Llama-3.1-8B", "endpoints_compatible", "region:us" ]	null	2025-05-02T09:33:29Z	--- base_model: meta-llama/Llama-3.1-8B library_name: transformers model_name: meta-llama-Llama-3.1-8B-pubmedqa-first-lora-4-0.0001-no-prompt-template tags: - generated_from_trainer - trl - sft licence: license --- # Model Card for meta-llama-Llama-3.1-8B-pubmedqa-first-lora-4-0.0001-no-prompt-template This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B). It has been trained using [TRL](https://github.com/huggingface/trl). ## Quick start ```python from transformers import pipeline question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?" generator = pipeline("text-generation", model="ASethi04/meta-llama-Llama-3.1-8B-pubmedqa-first-lora-4-0.0001-no-prompt-template", device="cuda") output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0] print(output["generated_text"]) ``` ## Training procedure [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/torchql-org/huggingface/runs/l5i0u4nr) This model was trained with SFT. ### Framework versions - TRL: 0.16.1 - Transformers: 4.51.2 - Pytorch: 2.6.0 - Datasets: 3.5.0 - Tokenizers: 0.21.1 ## Citations Cite TRL as: ```bibtex @misc{vonwerra2022trl, title = {{TRL: Transformer Reinforcement Learning}}, author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec}, year = 2020, journal = {GitHub repository}, publisher = {GitHub}, howpublished = {\url{https://github.com/huggingface/trl}} } ```
ma921/gpt2-large_h_dpo_imdb_noise40_epoch30	ma921	2025-05-02T11:07:22Z	0	0	transformers	[ "transformers", "safetensors", "gpt2", "text-generation", "generated_from_trainer", "base_model:ma921/gpt2-large-sft-imdb", "base_model:finetune:ma921/gpt2-large-sft-imdb", "license:mit", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T11:06:25Z	--- library_name: transformers license: mit base_model: ma921/gpt2-large-sft-imdb tags: - generated_from_trainer model-index: - name: gpt2-large_h_dpo_imdb_noise40_epoch30 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # gpt2-large_h_dpo_imdb_noise40_epoch30 This model is a fine-tuned version of [ma921/gpt2-large-sft-imdb](https://huggingface.co/ma921/gpt2-large-sft-imdb) on an unknown dataset. ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-06 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - distributed_type: multi-GPU - gradient_accumulation_steps: 32 - total_train_batch_size: 256 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 30 - mixed_precision_training: Native AMP ### Training results ### Framework versions - Transformers 4.51.3 - Pytorch 2.6.0+cu124 - Datasets 3.5.1 - Tokenizers 0.21.1
Sorawiz/Qwen2.5-14B-Instinct	Sorawiz	2025-05-02T11:05:45Z	16	0	transformers	[ "transformers", "safetensors", "qwen2", "text-generation", "mergekit", "merge", "conversational", "arxiv:2311.03099", "base_model:Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4", "base_model:merge:Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4", "base_model:Sao10K/14B-Qwen2.5-Freya-x1", "base_model:merge:Sao10K/14B-Qwen2.5-Freya-x1", "base_model:Sao10K/14B-Qwen2.5-Kunou-v1", "base_model:merge:Sao10K/14B-Qwen2.5-Kunou-v1", "base_model:SicariusSicariiStuff/Impish_QWEN_14B-1M", "base_model:merge:SicariusSicariiStuff/Impish_QWEN_14B-1M", "base_model:Sorawiz/Qwen2.5-14B-GCC", "base_model:merge:Sorawiz/Qwen2.5-14B-GCC", "base_model:Ttimofeyka/Tissint-14B-v1.2-128k-RP", "base_model:merge:Ttimofeyka/Tissint-14B-v1.2-128k-RP", "base_model:deepcogito/cogito-v1-preview-qwen-14B", "base_model:merge:deepcogito/cogito-v1-preview-qwen-14B", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-04-15T16:31:32Z	--- base_model: - Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4 - deepcogito/cogito-v1-preview-qwen-14B - Sao10K/14B-Qwen2.5-Freya-x1 - Sao10K/14B-Qwen2.5-Kunou-v1 - SicariusSicariiStuff/Impish_QWEN_14B-1M - Ttimofeyka/Tissint-14B-v1.2-128k-RP - Sorawiz/Qwen2.5-14B-GCC library_name: transformers tags: - mergekit - merge --- # Merge This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). ## Merge Details ### Merge Method This model was merged using the [DARE TIES](https://arxiv.org/abs/2311.03099) merge method using Sorawiz/Qwen2.5-14B-Imstinct as a base. ### Models Merged The following models were included in the merge: * Sorawiz/Qwen2.5-14B-1M-Instinct * [SicariusSicariiStuff/Impish_QWEN_14B-1M](https://huggingface.co/SicariusSicariiStuff/Impish_QWEN_14B-1M) * [Ttimofeyka/Tissint-14B-v1.2-128k-RP](https://huggingface.co/Ttimofeyka/Tissint-14B-v1.2-128k-RP) * [Sorawiz/Qwen2.5-14B-GCC](https://huggingface.co/Sorawiz/Qwen2.5-14B-GCC) ### Configuration The following YAML configuration was used to produce this model: ```yaml name: Sorawiz/Qwen2.5-14B-Instinct-Base merge_method: dare_ties base_model: Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4 models: - model: Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4 parameters: weight: 0.3 - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP parameters: weight: 0.7 parameters: density: 1 tokenizer: source: union chat_template: auto --- name: Sorawiz/Qwen2.5-14B-Instincto merge_method: dare_ties base_model: deepcogito/cogito-v1-preview-qwen-14B models: - model: deepcogito/cogito-v1-preview-qwen-14B parameters: weight: 0.4 - model: Sorawiz/Qwen2.5-14B-Instinct-Base parameters: weight: 0.3 - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP parameters: weight: 0.3 parameters: density: 0.5 tokenizer: source: union chat_template: auto --- name: Sorawiz/Qwen2.5-14B-Imstinct merge_method: dare_ties base_model: Sorawiz/Qwen2.5-14B-Instincto models: - model: Sorawiz/Qwen2.5-14B-Instincto parameters: weight: 0.6 - model: SicariusSicariiStuff/Impish_QWEN_14B-1M parameters: weight: 0.2 - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP parameters: weight: 0.2 parameters: density: 0.5 tokenizer: source: SicariusSicariiStuff/Impish_QWEN_14B-1M chat_template: auto --- name: Sorawiz/Qwen2.5-14B-Instinct-Base merge_method: dare_ties base_model: Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4 models: - model: Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4 parameters: weight: 0.3 - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP parameters: weight: 0.7 parameters: density: 1 tokenizer: source: union chat_template: auto --- name: Sorawiz/Qwen2.5-14B-Instincto merge_method: dare_ties base_model: deepcogito/cogito-v1-preview-qwen-14B models: - model: deepcogito/cogito-v1-preview-qwen-14B parameters: weight: 0.4 - model: Sorawiz/Qwen2.5-14B-Instinct-Base parameters: weight: 0.3 - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP parameters: weight: 0.3 parameters: density: 0.5 tokenizer: source: union chat_template: auto --- name: Sorawiz/Qwen2.5-14B-Kunousint merge_method: dare_ties base_model: Sao10K/14B-Qwen2.5-Kunou-v1 models: - model: Sao10K/14B-Qwen2.5-Kunou-v1 parameters: weight: 0.5 - model: Sorawiz/Qwen2.5-14B-Instincto parameters: weight: 0.3 - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP parameters: weight: 0.2 parameters: density: 0.5 tokenizer: source: union chat_template: auto --- name: Sorawiz/Qwen2.5-14B-Kunousint-1M merge_method: dare_ties base_model: Sorawiz/Qwen2.5-14B-Imstinct models: - model: Sorawiz/Qwen2.5-14B-Imstinct parameters: weight: 0.2 - model: Sorawiz/Qwen2.5-14B-Kunousint parameters: weight: 0.5 - model: Sao10K/14B-Qwen2.5-Kunou-v1 parameters: weight: 0.3 parameters: density: 0.5 tokenizer: source: union chat_template: auto --- name: Sorawiz/Qwen2.5-14B-Frayasint merge_method: dare_ties base_model: Sao10K/14B-Qwen2.5-Freya-x1 models: - model: Sao10K/14B-Qwen2.5-Freya-x1 parameters: weight: 0.5 - model: Sorawiz/Qwen2.5-14B-Instincto parameters: weight: 0.3 - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP parameters: weight: 0.2 parameters: density: 0.5 tokenizer: source: union chat_template: auto --- name: Sorawiz/Qwen2.5-14B-Frayasint-1M merge_method: dare_ties base_model: Sorawiz/Qwen2.5-14B-Imstinct models: - model: Sorawiz/Qwen2.5-14B-Imstinct parameters: weight: 0.2 - model: Sorawiz/Qwen2.5-14B-Frayasint parameters: weight: 0.5 - model: Sao10K/14B-Qwen2.5-Freya-x1 parameters: weight: 0.3 parameters: density: 0.5 tokenizer: source: union chat_template: auto --- name: Sorawiz/Qwen2.5-14B-1M-Instinct merge_method: dare_ties base_model: Sorawiz/Qwen2.5-14B-Imstinct models: - model: Sorawiz/Qwen2.5-14B-Imstinct parameters: weight: 0.25 - model: Sorawiz/Qwen2.5-14B-1M-Kunousint-1M parameters: weight: 0.25 - model: Sorawiz/Qwen2.5-14B-Frayasint-1M parameters: weight: 0.25 - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP parameters: weight: 0.25 parameters: density: 1 tokenizer: source: union chat_template: auto --- merge_method: dare_ties base_model: Sorawiz/Qwen2.5-14B-Imstinct models: - model: Sorawiz/Qwen2.5-14B-Imstinct parameters: weight: 0.20 - model: Sorawiz/Qwen2.5-14B-GCC parameters: weight: 0.20 - model: Sorawiz/Qwen2.5-14B-1M-Instinct parameters: weight: 0.20 - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP parameters: weight: 0.20 - model: SicariusSicariiStuff/Impish_QWEN_14B-1M parameters: weight: 0.20 parameters: density: 1 tokenizer: source: union chat_template: auto ```
naveennagar0909/lora-coke-can-flux-1dev	naveennagar0909	2025-05-02T11:05:11Z	0	0	diffusers	[ "diffusers", "text-to-image", "diffusers-training", "lora", "flux", "flux-diffusers", "template:sd-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "license:other", "region:us" ]	text-to-image	2025-05-02T10:59:05Z	--- base_model: black-forest-labs/FLUX.1-dev library_name: diffusers license: other instance_prompt: a photo of sks coke can widget: - text: A photo of sks coke can on a mountain output: url: image_0.png tags: - text-to-image - diffusers-training - diffusers - lora - flux - flux-diffusers - template:sd-lora --- <!-- This model card has been generated automatically according to the information the training script had access to. You should probably proofread and complete it, then remove this comment. --> # Flux DreamBooth LoRA - naveennagar0909/lora-coke-can-flux-1dev <Gallery /> ## Model description These are naveennagar0909/lora-coke-can-flux-1dev DreamBooth LoRA weights for black-forest-labs/FLUX.1-dev. The weights were trained using [DreamBooth](https://dreambooth.github.io/) with the [Flux diffusers trainer](https://github.com/huggingface/diffusers/blob/main/examples/dreambooth/README_flux.md). Was LoRA for the text encoder enabled? False. ## Trigger words You should use `a photo of sks coke can` to trigger the image generation. ## Download model [Download the *.safetensors LoRA](naveennagar0909/lora-coke-can-flux-1dev/tree/main) in the Files & versions tab. ## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers) ```py from diffusers import AutoPipelineForText2Image import torch pipeline = AutoPipelineForText2Image.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16).to('cuda') pipeline.load_lora_weights('naveennagar0909/lora-coke-can-flux-1dev', weight_name='pytorch_lora_weights.safetensors') image = pipeline('A photo of sks coke can on a mountain').images[0] ``` For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters) ## License Please adhere to the licensing terms as described [here](https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md). ## Intended uses & limitations #### How to use ```python # TODO: add an example code snippet for running this diffusion pipeline ``` #### Limitations and bias [TODO: provide examples of latent issues and potential remediations] ## Training details [TODO: describe the data used to train the model]
mradermacher/Meditron3-Qwen2.5-14B-GGUF	mradermacher	2025-05-02T11:04:40Z	160	1	transformers	[ "transformers", "gguf", "medical", "zho", "eng", "fra", "spa", "por", "deu", "ita", "rus", "jpn", "kor", "vie", "tha", "ara", "base_model:OpenMeditron/Meditron3-Qwen2.5-14B", "base_model:quantized:OpenMeditron/Meditron3-Qwen2.5-14B", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]	null	2025-02-19T06:40:06Z	--- base_model: OpenMeditron/Meditron3-Qwen2.5-14B language: - zho - eng - fra - spa - por - deu - ita - rus - jpn - kor - vie - tha - ara library_name: transformers license: apache-2.0 quantized_by: mradermacher tags: - medical --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/OpenMeditron/Meditron3-Qwen2.5-14B <!-- provided-files --> weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q2_K.gguf) \| Q2_K \| 5.9 \| \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q3_K_S.gguf) \| Q3_K_S \| 6.8 \| \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q3_K_M.gguf) \| Q3_K_M \| 7.4 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q3_K_L.gguf) \| Q3_K_L \| 8.0 \| \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.IQ4_XS.gguf) \| IQ4_XS \| 8.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q4_K_S.gguf) \| Q4_K_S \| 8.7 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q4_K_M.gguf) \| Q4_K_M \| 9.1 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q5_K_S.gguf) \| Q5_K_S \| 10.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q5_K_M.gguf) \| Q5_K_M \| 10.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q6_K.gguf) \| Q6_K \| 12.2 \| very good quality \| \| [GGUF](https://huggingface.co/mradermacher/Meditron3-Qwen2.5-14B-GGUF/resolve/main/Meditron3-Qwen2.5-14B.Q8_0.gguf) \| Q8_0 \| 15.8 \| fast, best quality \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
Kapil003/qwen2.5-1.5b-finetuned-conversation	Kapil003	2025-05-02T11:03:25Z	0	0	transformers	[ "transformers", "safetensors", "arxiv:1910.09700", "endpoints_compatible", "region:us" ]	null	2025-05-02T11:03:21Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
luckycanucky/hnd_model_16b-Q4_K_M-GGUF	luckycanucky	2025-05-02T11:03:14Z	0	0	transformers	[ "transformers", "gguf", "text-generation-inference", "unsloth", "llama", "trl", "sft", "llama-cpp", "gguf-my-repo", "en", "base_model:luckycanucky/hnd_model_16b", "base_model:quantized:luckycanucky/hnd_model_16b", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]	null	2025-05-02T11:03:02Z	--- base_model: luckycanucky/hnd_model_16b language: - en license: apache-2.0 tags: - text-generation-inference - transformers - unsloth - llama - trl - sft - llama-cpp - gguf-my-repo --- # luckycanucky/hnd_model_16b-Q4_K_M-GGUF This model was converted to GGUF format from [`luckycanucky/hnd_model_16b`](https://huggingface.co/luckycanucky/hnd_model_16b) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space. Refer to the [original model card](https://huggingface.co/luckycanucky/hnd_model_16b) for more details on the model. ## Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) ```bash brew install llama.cpp ``` Invoke the llama.cpp server or the CLI. ### CLI: ```bash llama-cli --hf-repo luckycanucky/hnd_model_16b-Q4_K_M-GGUF --hf-file hnd_model_16b-q4_k_m.gguf -p "The meaning to life and the universe is" ``` ### Server: ```bash llama-server --hf-repo luckycanucky/hnd_model_16b-Q4_K_M-GGUF --hf-file hnd_model_16b-q4_k_m.gguf -c 2048 ``` Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well. Step 1: Clone llama.cpp from GitHub. ``` git clone https://github.com/ggerganov/llama.cpp ``` Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux). ``` cd llama.cpp && LLAMA_CURL=1 make ``` Step 3: Run inference through the main binary. ``` ./llama-cli --hf-repo luckycanucky/hnd_model_16b-Q4_K_M-GGUF --hf-file hnd_model_16b-q4_k_m.gguf -p "The meaning to life and the universe is" ``` or ``` ./llama-server --hf-repo luckycanucky/hnd_model_16b-Q4_K_M-GGUF --hf-file hnd_model_16b-q4_k_m.gguf -c 2048 ```
Japheth11/Paddymonet	Japheth11	2025-05-02T11:02:36Z	0	0	null	[ "license:apache-2.0", "region:us" ]	null	2025-05-02T11:02:36Z	--- license: apache-2.0 ---
fax4ever/culturalitems-roberta-base-1	fax4ever	2025-05-02T11:01:57Z	0	0	null	[ "safetensors", "roberta", "license:apache-2.0", "region:us" ]	null	2025-05-02T10:18:55Z	--- license: apache-2.0 ---
luckycanucky/lora_model_hnd-F16-GGUF	luckycanucky	2025-05-02T11:01:10Z	0	0	transformers	[ "transformers", "gguf", "text-generation-inference", "unsloth", "llama", "trl", "llama-cpp", "gguf-my-lora", "en", "base_model:luckycanucky/lora_model_hnd", "base_model:quantized:luckycanucky/lora_model_hnd", "license:apache-2.0", "endpoints_compatible", "region:us" ]	null	2025-05-02T11:01:02Z	--- base_model: luckycanucky/lora_model_hnd tags: - text-generation-inference - transformers - unsloth - llama - trl - llama-cpp - gguf-my-lora license: apache-2.0 language: - en --- # luckycanucky/lora_model_hnd-F16-GGUF This LoRA adapter was converted to GGUF format from [`luckycanucky/lora_model_hnd`](https://huggingface.co/luckycanucky/lora_model_hnd) via the ggml.ai's [GGUF-my-lora](https://huggingface.co/spaces/ggml-org/gguf-my-lora) space. Refer to the [original adapter repository](https://huggingface.co/luckycanucky/lora_model_hnd) for more details. ## Use with llama.cpp ```bash # with cli llama-cli -m base_model.gguf --lora lora_model_hnd-f16.gguf (...other args) # with server llama-server -m base_model.gguf --lora lora_model_hnd-f16.gguf (...other args) ``` To know more about LoRA usage with llama.cpp server, refer to the [llama.cpp server documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/server/README.md).
hanaearg/emo-gemma2-9b-eng-10epochs	hanaearg	2025-05-02T10:59:02Z	0	0	transformers	[ "transformers", "safetensors", "text-generation-inference", "unsloth", "gemma2", "trl", "en", "base_model:unsloth/gemma-2-9b-it-bnb-4bit", "base_model:finetune:unsloth/gemma-2-9b-it-bnb-4bit", "license:apache-2.0", "endpoints_compatible", "region:us" ]	null	2025-05-02T10:58:51Z	--- base_model: unsloth/gemma-2-9b-it-bnb-4bit tags: - text-generation-inference - transformers - unsloth - gemma2 - trl license: apache-2.0 language: - en --- # Uploaded model - Developed by: hanaearg - License: apache-2.0 - Finetuned from model : unsloth/gemma-2-9b-it-bnb-4bit This gemma2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
luckycanucky/hnd_model_16b	luckycanucky	2025-05-02T10:52:44Z	0	0	transformers	[ "transformers", "pytorch", "llama", "text-generation", "text-generation-inference", "unsloth", "trl", "sft", "conversational", "en", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T10:50:32Z	--- base_model: unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - llama - trl - sft license: apache-2.0 language: - en --- # Uploaded model - Developed by: luckycanucky - License: apache-2.0 - Finetuned from model : unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
naveennagar0909/lora-bicycle-flux-1dev	naveennagar0909	2025-05-02T10:52:40Z	0	0	diffusers	[ "diffusers", "text-to-image", "diffusers-training", "lora", "flux", "flux-diffusers", "template:sd-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "license:other", "region:us" ]	text-to-image	2025-05-02T10:06:45Z	--- base_model: black-forest-labs/FLUX.1-dev library_name: diffusers license: other instance_prompt: a photo of sks bicycle widget: - text: A photo of sks bicycle on a mountain output: url: image_0.png tags: - text-to-image - diffusers-training - diffusers - lora - flux - flux-diffusers - template:sd-lora --- <!-- This model card has been generated automatically according to the information the training script had access to. You should probably proofread and complete it, then remove this comment. --> # Flux DreamBooth LoRA - naveennagar0909/lora-bicycle-flux-1dev <Gallery /> ## Model description These are naveennagar0909/lora-bicycle-flux-1dev DreamBooth LoRA weights for black-forest-labs/FLUX.1-dev. The weights were trained using [DreamBooth](https://dreambooth.github.io/) with the [Flux diffusers trainer](https://github.com/huggingface/diffusers/blob/main/examples/dreambooth/README_flux.md). Was LoRA for the text encoder enabled? False. ## Trigger words You should use `a photo of sks bicycle` to trigger the image generation. ## Download model [Download the *.safetensors LoRA](naveennagar0909/lora-bicycle-flux-1dev/tree/main) in the Files & versions tab. ## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers) ```py from diffusers import AutoPipelineForText2Image import torch pipeline = AutoPipelineForText2Image.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16).to('cuda') pipeline.load_lora_weights('naveennagar0909/lora-bicycle-flux-1dev', weight_name='pytorch_lora_weights.safetensors') image = pipeline('A photo of sks bicycle on a mountain').images[0] ``` For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters) ## License Please adhere to the licensing terms as described [here](https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md). ## Intended uses & limitations #### How to use ```python # TODO: add an example code snippet for running this diffusion pipeline ``` #### Limitations and bias [TODO: provide examples of latent issues and potential remediations] ## Training details [TODO: describe the data used to train the model]
hello231/gemma-qa-finetuned	hello231	2025-05-02T10:50:56Z	0	0	transformers	[ "transformers", "safetensors", "gemma", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "4-bit", "bitsandbytes", "region:us" ]	text-generation	2025-05-02T10:49:28Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
luckycanucky/lora_model_hnd	luckycanucky	2025-05-02T10:49:27Z	0	0	transformers	[ "transformers", "safetensors", "text-generation-inference", "unsloth", "llama", "trl", "en", "license:apache-2.0", "endpoints_compatible", "region:us" ]	null	2025-05-02T10:49:18Z	--- base_model: unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - llama - trl license: apache-2.0 language: - en --- # Uploaded model - Developed by: luckycanucky - License: apache-2.0 - Finetuned from model : unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
mradermacher/legml-v0.1-GGUF	mradermacher	2025-05-02T10:47:16Z	36	1	transformers	[ "transformers", "gguf", "zho", "eng", "fra", "spa", "por", "deu", "ita", "rus", "jpn", "kor", "vie", "tha", "ara", "base_model:legmlai/legml-v0.1", "base_model:quantized:legmlai/legml-v0.1", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]	null	2025-03-01T12:18:13Z	--- base_model: legmlai/legml-v0.1 language: - zho - eng - fra - spa - por - deu - ita - rus - jpn - kor - vie - tha - ara library_name: transformers license: apache-2.0 quantized_by: mradermacher --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/legmlai/legml-v0.1 <!-- provided-files --> weighted/imatrix quants are available at https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q2_K.gguf) \| Q2_K \| 5.9 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q3_K_S.gguf) \| Q3_K_S \| 6.8 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q3_K_M.gguf) \| Q3_K_M \| 7.4 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q3_K_L.gguf) \| Q3_K_L \| 8.0 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.IQ4_XS.gguf) \| IQ4_XS \| 8.3 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q4_K_S.gguf) \| Q4_K_S \| 8.7 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q4_K_M.gguf) \| Q4_K_M \| 9.1 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q5_K_S.gguf) \| Q5_K_S \| 10.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q5_K_M.gguf) \| Q5_K_M \| 10.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q6_K.gguf) \| Q6_K \| 12.2 \| very good quality \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-GGUF/resolve/main/legml-v0.1.Q8_0.gguf) \| Q8_0 \| 15.8 \| fast, best quality \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to. <!-- end -->
mradermacher/legml-v0.1-i1-GGUF	mradermacher	2025-05-02T10:47:07Z	313	1	transformers	[ "transformers", "gguf", "zho", "eng", "fra", "spa", "por", "deu", "ita", "rus", "jpn", "kor", "vie", "tha", "ara", "base_model:legmlai/legml-v0.1", "base_model:quantized:legmlai/legml-v0.1", "license:apache-2.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	null	2025-03-01T13:51:51Z	--- base_model: legmlai/legml-v0.1 language: - zho - eng - fra - spa - por - deu - ita - rus - jpn - kor - vie - tha - ara library_name: transformers license: apache-2.0 quantized_by: mradermacher --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: nicoboss --> weighted/imatrix quants of https://huggingface.co/legmlai/legml-v0.1 <!-- provided-files --> static quants are available at https://huggingface.co/mradermacher/legml-v0.1-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) \| Link \| Type \| Size/GB \| Notes \| \|:-----\|:-----\|--------:\|:------\| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ1_S.gguf) \| i1-IQ1_S \| 3.7 \| for the desperate \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ1_M.gguf) \| i1-IQ1_M \| 4.0 \| mostly desperate \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ2_XXS.gguf) \| i1-IQ2_XXS \| 4.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ2_XS.gguf) \| i1-IQ2_XS \| 4.8 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ2_S.gguf) \| i1-IQ2_S \| 5.1 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ2_M.gguf) \| i1-IQ2_M \| 5.5 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q2_K_S.gguf) \| i1-Q2_K_S \| 5.5 \| very low quality \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q2_K.gguf) \| i1-Q2_K \| 5.9 \| IQ3_XXS probably better \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ3_XXS.gguf) \| i1-IQ3_XXS \| 6.0 \| lower quality \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ3_XS.gguf) \| i1-IQ3_XS \| 6.5 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q3_K_S.gguf) \| i1-Q3_K_S \| 6.8 \| IQ3_XS probably better \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ3_S.gguf) \| i1-IQ3_S \| 6.8 \| beats Q3_K* \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ3_M.gguf) \| i1-IQ3_M \| 7.0 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q3_K_M.gguf) \| i1-Q3_K_M \| 7.4 \| IQ3_S probably better \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q3_K_L.gguf) \| i1-Q3_K_L \| 8.0 \| IQ3_M probably better \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ4_XS.gguf) \| i1-IQ4_XS \| 8.2 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q4_0.gguf) \| i1-Q4_0 \| 8.6 \| fast, low quality \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-IQ4_NL.gguf) \| i1-IQ4_NL \| 8.6 \| prefer IQ4_XS \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q4_K_S.gguf) \| i1-Q4_K_S \| 8.7 \| optimal size/speed/quality \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q4_K_M.gguf) \| i1-Q4_K_M \| 9.1 \| fast, recommended \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q4_1.gguf) \| i1-Q4_1 \| 9.5 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q5_K_S.gguf) \| i1-Q5_K_S \| 10.4 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q5_K_M.gguf) \| i1-Q5_K_M \| 10.6 \| \| \| [GGUF](https://huggingface.co/mradermacher/legml-v0.1-i1-GGUF/resolve/main/legml-v0.1.i1-Q6_K.gguf) \| i1-Q6_K \| 12.2 \| practically like static Q6_K \| Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to. <!-- end -->
hiltonv/gemma-3-finetune	hiltonv	2025-05-02T10:44:15Z	0	0	transformers	[ "transformers", "gemma3_text", "text-generation", "text-generation-inference", "unsloth", "gemma3", "conversational", "en", "base_model:unsloth/gemma-3-4b-it-unsloth-bnb-4bit", "base_model:finetune:unsloth/gemma-3-4b-it-unsloth-bnb-4bit", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T10:44:04Z	--- base_model: unsloth/gemma-3-4b-it-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - gemma3 license: apache-2.0 language: - en --- # Uploaded finetuned model - Developed by: hiltonv - License: apache-2.0 - Finetuned from model : unsloth/gemma-3-4b-it-unsloth-bnb-4bit This gemma3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
tattootryai/tattoo-pro-lora	tattootryai	2025-05-02T10:42:55Z	0	0	null	[ "license:other", "region:us" ]	null	2025-04-24T05:00:32Z	--- license: other license_name: openrail license_link: LICENSE ---
vertings6/12a8f32b-e0a5-4fcd-9ce6-5a92c2fd098d	vertings6	2025-05-02T10:39:32Z	0	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:unsloth/SmolLM-360M", "base_model:adapter:unsloth/SmolLM-360M", "license:apache-2.0", "4-bit", "bitsandbytes", "region:us" ]	null	2025-05-02T10:34:26Z	--- library_name: peft license: apache-2.0 base_model: unsloth/SmolLM-360M tags: - axolotl - generated_from_trainer model-index: - name: 12a8f32b-e0a5-4fcd-9ce6-5a92c2fd098d results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml absolute_data_files: true adapter: lora base_model: unsloth/SmolLM-360M bf16: true chat_template: llama3 dataset_prepared_path: /workspace/axolotl datasets: - data_files: - cadcf05882c29e35_train_data.json ds_type: json format: custom path: /workspace/input_data/cadcf05882c29e35_train_data.json type: field_input: system_prompt field_instruction: question field_output: response format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 144 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 2 gradient_checkpointing: true gradient_clipping: 0.5 group_by_length: false hub_model_id: vertings6/12a8f32b-e0a5-4fcd-9ce6-5a92c2fd098d hub_repo: null hub_strategy: end hub_token: null learning_rate: 3.0e-06 load_in_4bit: true load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 64 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 32 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 4 mixed_precision: bf16 mlflow_experiment_name: /tmp/cadcf05882c29e35_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 2048 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: ca58130f-99a5-458c-8fdf-cce4ae2a30fb wandb_project: s56-32 wandb_run: your_name wandb_runid: ca58130f-99a5-458c-8fdf-cce4ae2a30fb warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 12a8f32b-e0a5-4fcd-9ce6-5a92c2fd098d This model is a fine-tuned version of [unsloth/SmolLM-360M](https://huggingface.co/unsloth/SmolLM-360M) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.5955 ## 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: 3e-06 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 8 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 2.537 \| 0.0169 \| 200 \| 2.5955 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
kmcs-casulit/ts_ticket	kmcs-casulit	2025-05-02T10:37:33Z	0	0	transformers	[ "transformers", "safetensors", "mistral", "text-generation", "text-generation-inference", "unsloth", "trl", "conversational", "en", "base_model:unsloth/mistral-7b-instruct-v0.3-bnb-4bit", "base_model:finetune:unsloth/mistral-7b-instruct-v0.3-bnb-4bit", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T10:31:19Z	--- base_model: unsloth/mistral-7b-instruct-v0.3-bnb-4bit tags: - text-generation-inference - transformers - unsloth - mistral - trl license: apache-2.0 language: - en --- # Uploaded model - Developed by: kmcs-casulit - License: apache-2.0 - Finetuned from model : unsloth/mistral-7b-instruct-v0.3-bnb-4bit This mistral model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
Jathushan/TamilPaattu_bert	Jathushan	2025-05-02T10:37:16Z	0	0	transformers	[ "transformers", "safetensors", "bert", "fill-mask", "arxiv:1910.09700", "autotrain_compatible", "endpoints_compatible", "region:us" ]	fill-mask	2025-05-02T10:36:44Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
BABYSHARK09/Uni_6x7	BABYSHARK09	2025-05-02T10:34:46Z	0	0	transformers	[ "transformers", "safetensors", "llama", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T10:13:41Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
Bisher/train_run-qwen2.5-0.5b-instruct-arabic-diacritization-fadel-10k-left	Bisher	2025-05-02T10:32:01Z	0	0	peft	[ "peft", "safetensors", "generated_from_trainer", "base_model:Qwen/Qwen2.5-0.5B-Instruct", "base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct", "license:apache-2.0", "region:us" ]	null	2025-05-02T09:03:58Z	--- library_name: peft license: apache-2.0 base_model: Qwen/Qwen2.5-0.5B-Instruct tags: - generated_from_trainer model-index: - name: train_run-qwen2.5-0.5b-instruct-arabic-diacritization-fadel-10k-left results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # train_run-qwen2.5-0.5b-instruct-arabic-diacritization-fadel-10k-left This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.1130 ## 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: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.PAGED_ADAMW with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: constant - lr_scheduler_warmup_ratio: 0.03 - num_epochs: 1 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:-----:\|:----:\|:---------------:\| \| 0.2675 \| 0.16 \| 100 \| 0.2502 \| \| 0.1788 \| 0.32 \| 200 \| 0.1742 \| \| 0.1496 \| 0.48 \| 300 \| 0.1526 \| \| 0.1349 \| 0.64 \| 400 \| 0.1319 \| \| 0.1195 \| 0.8 \| 500 \| 0.1194 \| \| 0.1105 \| 0.96 \| 600 \| 0.1130 \| ### Framework versions - PEFT 0.14.0 - Transformers 4.51.1 - Pytorch 2.5.1+cu124 - Datasets 3.5.0 - Tokenizers 0.21.0
akhilfau/smollm2-360m-physics-gguf	akhilfau	2025-05-02T10:29:07Z	13	0	null	[ "gguf", "endpoints_compatible", "region:us", "conversational" ]	null	2025-04-15T20:15:48Z	smollm2-360m-physics-gguf Author: [Akhil Vallala](https://www.linkedin.com/in/akhil-fau) Base Model: [`akhilfau/fine-tuned-smolLM2-360M-with-on-combined_Instruction_dataset`](https://huggingface.co/akhilfau/fine-tuned-smolLM2-360M-with-on-combined_Instruction_dataset) Architecture: LLaMA (SmolLM2) Parameter count: 362M Format: GGUF (Q4_K_M, Q8_0, FP16) License: Apache 2.0 Model Type: Instruction-tuned Small Language Model (SLM) Use Case: Solving physics word problems on mobile devices Model Overview This GGUF model is a quantized version of the Tiny-Physics model, based on SmolLM2-360M and fine-tuned for physics word problem solving using both real and synthetic datasets. It is designed to deliver accurate, low-latency performance on mobile and edge devices. Datasets Used - 📘 camel-ai/physics: Publicly available dataset with 20,000+ physics QA pairs - 📘 Seed dataset: Extracted from 1000 Solved Problems in Classical Physics - 🧠 Synthetic dataset: 6,279 rigorously validated question-answer pairs generated using a GPT-4o-based multi-agent system These datasets were formatted for instruction tuning using structured prompt–response pairs. Training Details - Model: SmolLM2-360M - Fine-tuning: Instruction fine-tuning with LoRA (Low-Rank Adaptation) - Libraries: Hugging Face Transformers, TRL, Lighteval - Training Epochs: 3 (best accuracy observed at 3–5 epochs) - Fine-tuning Objective: Maximize performance on MMLU College Physics - Best Model Accuracy: `24.51%` on MMLU College Physics Evaluation Evaluated with: [Lighteval](https://github.com/huggingface/lighteval) Benchmark: [MMLU College Physics](https://huggingface.co/datasets/hendrycks_test) Performance: \| Dataset \| Accuracy (SmolLM2-360M-Instruct) \| \|-----------------------------\|----------------------------------\| \| MMLU: College Physics \| 24.51% \| \| Instruction-Tuned camel-ai \| 25.49% \| \| Combined Instruction Dataset\| 24.51% \| GGUF Quantization Model is provided in multiple quantization formats: \| Format \| Size \| Accuracy Retention \| Inference Speed \| RAM Usage \| Target Use \| \|----------\|--------\|--------------------\|------------------\|---------------\|------------------------------------\| \| `Q4_K_M` \| ~271MB \| ~95–97% \| Fast \| ~600–800MB \| Ideal for mid-range mobile devices \| \| `Q8_0` \| ~386MB \| ~99% \| Medium \| ~1–1.5GB \| Best for higher-end devices \| \| `FP16` \| ~800MB \| 100% \| Slow \| ~2GB+ \| Reference use only \| How to Use ```bash # Using llama.cpp ./main -m smollm2-360m-physics-gguf.Q4_K_M.gguf -p "What is the acceleration of a 2kg mass falling from 5 meters?" ``` Or via `llama-cpp-python`: ```python from llama_cpp import Llama llm = Llama(model_path="smollm2-360m-physics-gguf.Q4_K_M.gguf") output = llm("What is the potential energy of a 3kg object at 10 meters?") ``` Intended Use - 📚 Physics tutoring apps - 📶 Offline mobile inference - 🧑‍🏫 Educational tools for conceptual reasoning - 🔋 Low-power deployment scenarios Limitations - Not trained on multiple-choice formatted data (MCQ output mismatch possible) - Topic imbalance in datasets may affect generalization - Not suitable for non-physics or open-domain tasks Carbon Footprint Training and fine-tuning consumed approx. 2.64 kg CO₂e, equivalent to a ~7-mile car ride. This was achieved using local GPU resources (RTX A5500) and energy-efficient batch tuning with LoRA. Citation ```bibtex @misc{vallala2025tinyphysics, title={Tiny-Physics: A Compact Large Language Model for Physics Word Problems on Mobile Devices}, author={Akhil Vallala}, year={2025}, howpublished={\url{https://huggingface.co/akhilfau/smollm2-360m-physics-gguf}}, } ```
Maxim01/Intent_Classification_Test	Maxim01	2025-05-02T10:27:50Z	13	0	setfit	[ "setfit", "safetensors", "bert", "sentence-transformers", "text-classification", "generated_from_setfit_trainer", "arxiv:2209.11055", "base_model:cointegrated/rubert-tiny2", "base_model:finetune:cointegrated/rubert-tiny2", "model-index", "region:us" ]	text-classification	2025-01-26T05:56:28Z	--- tags: - setfit - sentence-transformers - text-classification - generated_from_setfit_trainer widget: - text: Военная кафедра - text: Какие льготы есть у выпускников колледжа? - text: Какие этапы включает приемная кампания? - text: Сколько продлится приемная кампания - text: Какие требования для поступления в ВУЦ? metrics: - accuracy pipeline_tag: text-classification library_name: setfit inference: true base_model: cointegrated/rubert-tiny2 model-index: - name: SetFit with cointegrated/rubert-tiny2 results: - task: type: text-classification name: Text Classification dataset: name: Unknown type: unknown split: test metrics: - type: accuracy value: 1.0 name: Accuracy --- # SetFit with cointegrated/rubert-tiny2 This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [cointegrated/rubert-tiny2](https://huggingface.co/cointegrated/rubert-tiny2) as the Sentence Transformer embedding model. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification. The model has been trained using an efficient few-shot learning technique that involves: 1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning. 2. Training a classification head with features from the fine-tuned Sentence Transformer. ## Model Details ### Model Description - Model Type: SetFit - Sentence Transformer body: [cointegrated/rubert-tiny2](https://huggingface.co/cointegrated/rubert-tiny2) - Classification head: a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance - Maximum Sequence Length: 2048 tokens - Number of Classes: 7 classes <!-- - Training Dataset: [Unknown](https://huggingface.co/datasets/unknown) --> <!-- - Language: Unknown --> <!-- - License: Unknown --> ### Model Sources - Repository: [SetFit on GitHub](https://github.com/huggingface/setfit) - Paper: [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055) - Blogpost: [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit) ### Model Labels \| Label \| Examples \| \|:------\|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\| \| 3 \| <ul><li>'Есть ли военная кафедра'</li><li>'Какие документы нужны для поступления в ВУЦ?'</li><li>'Как проходит отбор на военную кафедру'</li></ul> \| \| 2 \| <ul><li>'Как подтвердить свои достижения?'</li><li>'Расскажи о дополнительных баллах при поступлении'</li><li>'Можно ли получить баллы за волонтерскую деятельность?'</li></ul> \| \| 1 \| <ul><li>'Нужно ли заверять копии документов?'</li><li>'Как узнать, что мои документы приняты?'</li><li>'Можно ли подать документы по почте?'</li></ul> \| \| 0 \| <ul><li>'Как долго длится приемная кампания'</li><li>'Продолжительность приемной кампании'</li><li>'Когда заканчивается приемная кампания?'</li></ul> \| \| 4 \| <ul><li>'Какие документы нужны для поступления после колледжа?'</li><li>'Как поступить в Вуз после колледжа'</li><li>'Какие сроки подачи документов после колледжа?'</li></ul> \| \| 6 \| <ul><li>'Как подать заявление на общежитие?'</li><li>'Кто претендует на предоставление общежития'</li><li>'Как оплачивать проживание в общежитии?'</li></ul> \| \| 5 \| <ul><li>'Какие ошибки чаще всего допускают на экзаменах?'</li><li>'Какие вступительные испытания нужно сдавать?'</li><li>'Как подготовиться к вступительным экзаменам?'</li></ul> \| ## Evaluation ### Metrics \| Label \| Accuracy \| \|:--------\|:---------\| \| all \| 1.0 \| ## Uses ### Direct Use for Inference First install the SetFit library: ```bash pip install setfit ``` Then you can load this model and run inference. ```python from setfit import SetFitModel # Download from the 🤗 Hub model = SetFitModel.from_pretrained("Maxim01/Intent_Classification_Test") # Run inference preds = model("Военная кафедра") ``` <!-- ### Downstream Use List how someone could finetune this model on their own dataset. --> <!-- ### Out-of-Scope Use List how the model may foreseeably be misused and address what users ought not to do with the model. --> <!-- ## Bias, Risks and Limitations What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model. --> <!-- ### Recommendations What are recommendations with respect to the foreseeable issues? For example, filtering explicit content. --> ## Training Details ### Training Set Metrics \| Training set \| Min \| Median \| Max \| \|:-------------\|:----\|:-------\|:----\| \| Word count \| 2 \| 5.3333 \| 8 \| \| Label \| Training Sample Count \| \|:------\|:----------------------\| \| 0 \| 9 \| \| 1 \| 11 \| \| 2 \| 6 \| \| 3 \| 9 \| \| 4 \| 10 \| \| 5 \| 11 \| \| 6 \| 10 \| ### Training Hyperparameters - batch_size: (8, 8) - num_epochs: (1, 1) - max_steps: -1 - sampling_strategy: oversampling - num_iterations: 20 - body_learning_rate: (2e-05, 2e-05) - head_learning_rate: 2e-05 - loss: CosineSimilarityLoss - distance_metric: cosine_distance - margin: 0.25 - end_to_end: False - use_amp: False - warmup_proportion: 0.1 - l2_weight: 0.01 - seed: 42 - eval_max_steps: -1 - load_best_model_at_end: False ### Training Results \| Epoch \| Step \| Training Loss \| Validation Loss \| \|:------:\|:----:\|:-------------:\|:---------------:\| \| 0.0030 \| 1 \| 0.1199 \| - \| \| 0.1515 \| 50 \| 0.1727 \| - \| \| 0.3030 \| 100 \| 0.0936 \| - \| \| 0.4545 \| 150 \| 0.0599 \| - \| \| 0.6061 \| 200 \| 0.0529 \| - \| \| 0.7576 \| 250 \| 0.0436 \| - \| \| 0.9091 \| 300 \| 0.0359 \| - \| ### Framework Versions - Python: 3.11.12 - SetFit: 1.1.2 - Sentence Transformers: 3.4.1 - Transformers: 4.51.3 - PyTorch: 2.6.0+cu124 - Datasets: 3.5.1 - Tokenizers: 0.21.1 ## Citation ### BibTeX ```bibtex @article{https://doi.org/10.48550/arxiv.2209.11055, doi = {10.48550/ARXIV.2209.11055}, url = {https://arxiv.org/abs/2209.11055}, author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren}, keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences}, title = {Efficient Few-Shot Learning Without Prompts}, publisher = {arXiv}, year = {2022}, copyright = {Creative Commons Attribution 4.0 International} } ``` <!-- ## Glossary Clearly define terms in order to be accessible across audiences. --> <!-- ## Model Card Authors Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction. --> <!-- ## Model Card Contact Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors. -->
basilepp19/bloom-1b7-evalita	basilepp19	2025-05-02T10:27:24Z	7	0	transformers	[ "transformers", "pytorch", "safetensors", "bloom", "text-generation", "it", "license:bigscience-bloom-rail-1.0", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2023-10-05T15:31:06Z	--- license: bigscience-bloom-rail-1.0 language: - it --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> This model is obtained by fine-tuning the BLOOM model over two Italian classification task prompts without language adaptation. To deal with this step, we decided to use data from two well-known EVALITA tasks: AMI2020 (misogyny detection) and HASPEEDE-v2-2020 (hate-speech detection). ## Model Details ### Model Description The BLOOM model is directly fine-tuned over two Italian classification task prompts using two well-known EVALITA tasks: AMI2020 (misogyny detection) and HASPEEDE-v2-2020 (hate-speech detection). We transformed the training data of the two tasks into an LLM prompt following a template. For the AMI task, we used the following template: instruction: Nel testo seguente si esprime odio contro le donne? Rispondi sì o no., input: \<text\>, output: \<sì/no\>. Similarly, for HASPEEDE we used: instruction: “Il testo seguente incita all’odio? Rispondi sì o no., input: \<text\>, output: \<sì/no\>. To fill these templates, we mapped the label "1" with the word "sì" and the label "0" with the word "no", \<text\> is just the sentence from the dataset to classify. To fine-tune the model, we use the script available here: https://github.com/hyintell/BLOOM-fine-tuning/tree/main - Developed by: Pierpaolo Basile, Pierluigi Cassotti, Marco Polignano, Lucia Siciliani, Giovanni Semeraro. Department of Computer Science, University of Bari Aldo Moro, Italy - Model type: BLOOM - Language(s) (NLP): Italian - License: BigScience BLOOM RAIL 1.0 ## Citation Pierpaolo Basile, Pierluigi Cassotti, Marco Polignano, Lucia Siciliani, Giovanni Semeraro. On the impact of Language Adaptation for Large Language Models: A case study for the Italian language using only open resources. Proceedings of the Ninth Italian Conference on Computational Linguistics (CLiC-it 2023).
gsingh1-py/llama3.2_1b_alignguard	gsingh1-py	2025-05-02T10:27:06Z	0	0	peft	[ "peft", "safetensors", "arxiv:1910.09700", "base_model:TinyLlama/TinyLlama-1.1B-Chat-v1.0", "base_model:adapter:TinyLlama/TinyLlama-1.1B-Chat-v1.0", "region:us" ]	null	2025-05-02T10:26:16Z	--- base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0 library_name: peft --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed] ### Framework versions - PEFT 0.15.2
BABYSHARK09/Uni_6x4	BABYSHARK09	2025-05-02T10:26:33Z	0	0	transformers	[ "transformers", "safetensors", "llama", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T10:13:22Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
BABYSHARK09/Uni_6x3	BABYSHARK09	2025-05-02T10:26:26Z	0	0	transformers	[ "transformers", "safetensors", "llama", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T10:13:17Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
Kevinvbezanson/SupraKetoBHBACVGummiesUS	Kevinvbezanson	2025-05-02T10:25:44Z	0	0	null	[ "region:us" ]	null	2025-05-02T10:25:04Z	<h2 style="text-align: center;"><span style="background-color: #ccffcc; color: #333399;"><em><strong>Buy Now :</strong></em></span></h2> <h2 style="text-align: center;"><span style="background-color: #ffffff;"><strong><a style="background-color: #ffffff;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>https://www.facebook.com/SupraKETOBHBACVGummiesUS/</strong></a> </strong></span></h2> <h3><strong><span style="color: #ff0000;">➢ Product Name</span> :</strong><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong> Supra Keto BHB + ACV Gummies</strong></a></h3> <h3><strong><span style="color: #808000;">➢ Main Benefits</span> :</strong><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong> Weight Loss, Health Benefit</strong></a></h3> <h3><strong><span style="color: #800080;">➢ Composition</span> :</strong><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong> Natural Organic Compound</strong></a></h3> <h3><strong><span style="background-color: #ffffff; color: #808080;">➢ Side-Effects</span>:</strong><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong> NA</strong></a></h3> <h3><strong><span style="color: #ff6600;">➢ Rating</span> :</strong><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong>⭐⭐⭐⭐⭐</strong></a></h3> <h3><strong><span style="color: #ff00ff;">➢ Availability</span> :</strong><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong> Online</strong></a></h3> <h3><strong><span style="color: #008080;">➢ Where to Buy</span> :</strong><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong> Click Here to Rush Your Order from the Official Website</strong></a></h3> <p> </p> <h2 style="text-align: center;"><span style="color: #993300;"><a style="color: #993300;" href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong>✅Visit The Official Website To Get Your Bottle Now✅</strong></a></span></h2> <h2 style="text-align: center;"><span style="color: #ff9900;"><a style="color: #ff9900;" href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong>✅Visit The Official Website To Get Your Bottle Now✅</strong></a></span></h2> <h2 style="text-align: center;"><span style="color: #993300;"><a style="color: #993300;" href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong>✅Visit The Official Website To Get Your Bottle Now✅</strong></a></span></h2> <p><span style="font-weight: 400;"> </span></p> <p><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> are one item that has recently attracted attention. Promoted as a powerful fat-burning supplement, these gummies are designed to help users transition into ketosis.  The body burns fat for energy instead of carbohydrates in this efficient metabolic state.  People who are trying to lose weight often run into a number of obstacles, such as the constant need for sugary snacks and the fatigue that comes with restricting calories. Enter Supra Keto BHB + ACV Gummies Australia, which claim to provide a practical and tasty solution for these common issues.</span></p> <p><span style="font-weight: 400;">A product like this has a lot of appeal not only because it can help you lose weight but also because it says it will make you feel better in general. Packed with natural ingredients, these gummies suggest a method of fat burning that aligns with the ketogenic lifestyle without the need for strict dieting or exhausting workouts.  They seem perfectly suited for those looking to incorporate a supplement into their daily routine while indulging in a flavorful treat.</span></p> <p><span style="font-weight: 400;">This comprehensive review will explore</span> <span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> in detail—from understanding their ingredients and benefits to pricing and availability.  We want to give you an objective opinion on whether or not these gummies can help you lose weight. So, if you’re curious about how Supra Keto BHB + ACV Gummies work and whether they can help you achieve your fitness goals, read on to learn more.</span></p> <p> </p> <p style="text-align: center;"><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/ "><img src="https://i.ibb.co/YFrW41tN/Supra-Keto-BHB-ACV-Gummies-6.jpg" alt="Supra-Keto-BHB-ACV-Gummies-6" border="0" /></a></p> <p> </p> <h1 style="text-align: center;"><em><span style="color: #ff0000; background-color: #ffcc99;"><a style="color: #ff0000; background-color: #ffcc99;" href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong>👉Supra Keto BHB + ACV Gummies For Special Discounts – Official Website 👈</strong></a></span></em></h1> <p><span style="font-weight: 400;"> </span></p> <h3><strong>What is Supra Keto BHB + ACV Gummies US?</strong></h3> <p><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> are a dietary supplement specifically formulated for individuals pursuing weight loss through the ketogenic diet.  The ketogenic diet emphasizes a drastic reduction in carbohydrate intake, which prompts the body to enter a state known as ketosis.  Ketosis is a good way for many people to lose weight because the body switches from using carbohydrates as its primary energy source to burning stored fats. These gummies make entering ketosis easier and faster.  By incorporating a blend of natural ingredients, Supra Keto BHB + ACV Gummies are designed to support fat burning, enhance energy levels, and promote overall well-being.  Their convenient form sets these gummies apart from traditional weight loss supplements.  Taking multiple pills throughout the day can be difficult for many people; these tasty gummies provide an easy and enjoyable alternative. Supra Keto BHB + ACV Gummies Australia' original formulation aims to help people who struggle with the typical carb-heavy diet, which makes losing weight nearly impossible. Without resorting to strict dieting or strenuous exercise, users of these gummies may be able to maintain higher energy levels, curb cravings, and reduce their appetite. This aspect makes them an attractive option for those who wish to enhance their weight loss journey while enjoying the process.</span></p> <h3><strong> Does Supra Keto BHB + ACV Gummies Work?</strong></h3> <p><span style="font-weight: 400;">The effectiveness of</span> <span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> largely hinges on their ability to assist the body in entering and maintaining ketosis.  As the body transitions into this metabolic state, it utilizes fat stores for energy, leading to weight loss.  Many users have reported positive results, including significant fat loss within a few weeks of consistent use.</span></p> <p><span style="font-weight: 400;">However, it’s essential to note that while these gummies can facilitate ketosis, optimal results are often best achieved when combined with a balanced diet and regular exercise.  The gummies can boost your body's ability to burn fat by providing the right conditions, allowing you to lose the most weight over time. Moreover, numerous testimonials and reviews suggest that users have experienced increased energy levels, improved mental clarity, and reduced appetite while using the product.  These additional benefits support a weight loss regimen and enhance overall quality of life, making it easier to stay committed to healthy habits.</span></p> <p><span style="font-weight: 400;"> While individual experiences may vary, it is evident that</span> <span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> can support your weight loss journey.  They complement keto diets and lifestyle changes in a practical and enjoyable way, bringing you closer to your goals while making the process more manageable and less intimidating.</span></p> <p> </p> <p style="text-align: center;"><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/ "><img src="https://i.ibb.co/kg8Z3c8h/Supra-Keto-BHB-ACV-Gummies-4.png" alt="Supra-Keto-BHB-ACV-Gummies-4" border="0" /></a></p> <p> </p> <h1 style="text-align: center;"><span style="background-color: #ffcc99; color: #008000;"><em><a style="background-color: #ffcc99; color: #008000;" href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong>👉(Exceptional Proposition) Supra Keto BHB + ACV Gummies For Weight reduction Official Site 👈</strong></a></em></span></h1> <h3><strong> What are the ingredients in Supra Keto BHB + ACV Gummies US?</strong></h3> <p><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span> <span style="font-weight: 400;">boast a carefully selected blend of high-quality ingredients that synergistically promote weight loss and overall health.  The primary components of the formula are listed below, each chosen for their distinct advantages. </span></p> <p><span style="font-weight: 400;">The liver produces beta-hydroxybutyrate (BHB), a naturally occurring ketone, during periods of fasting or restriction of carbohydrate intake. It serves as an alternative energy source for the brain and body when glucose levels drop.  In the context of Supra Keto BHB + ACV Gummies Australia, BHB plays a crucial role in helping the body enter ketosis more rapidly.  By providing an exogenous source of ketones through the gummies, BHB can increase fat oxidation, physical performance, and energy levels. The transition into a ketogenic state may be made easier and more effective by users' increased mental clarity and decreased fatigue. Ceylon Cinnamon Bark</span></p> <p><span style="font-weight: 400;"> Cinnamon from Ceylon is well-known for its many health benefits, especially its ability to control blood sugar levels. Including this ingredient in</span> <span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> may promote better appetite control, decrease cravings, and enhance the body’s sensitivity to insulin.  Cinnamon is also known for its antioxidant properties, which can contribute to overall well-being by fighting inflammation and oxidative stress.  This ingredient adds a delightful flavor to the gummies and supports the metabolic processes involved in weight loss.</span></p> <ul> <li> <h4><span style="font-weight: 400;"> </span><strong>Corosolic Acid (from Banaba Leaf Extract)</strong></h4> </li> </ul> <p><span style="font-weight: 400;"> It has been demonstrated that the Banaba tree's leaves contain corosolic acid, which has potent anti-diabetic properties. It can help people who are trying to lose weight by lowering blood sugar levels and increasing their sensitivity to insulin. By promoting better glucose metabolism, Corosolic acid can help stabilize energy levels and prevent sugar cravings, making it easier to adhere to a ketogenic diet.  This ingredient enhances the overall efficacy of the gummies by supporting metabolic health alongside weight loss efforts.</span></p> <p><span style="font-weight: 400;"> </span></p> <ul> <li> <h4><strong>Bulk Saffron Extract Saffron</strong></h4> </li> </ul> <p><span style="font-weight: 400;">Bulk Saffron Extract Saffron is often celebrated not only for its culinary uses but also for its mood-enhancing properties.  Research suggests that saffron extract may help reduce emotional eating and improve overall mood, making it easier for individuals to resist unhealthy snacks.  Furthermore, saffron is rich in antioxidants, which can support mental health and wellness.  Supra Keto BHB + ACV Gummies may make it easier for users to stay motivated and positive throughout their weight loss journey by including saffron bulb extract. Fucoxanthin, which comes from Laminaria Japonica Fucoxanthin is a carotenoid found in brown seaweed, particularly Laminaria japonica.  Because of its potential to promote fat loss, particularly visceral fat, this compound has attracted attention. Fucoxanthin may increase thermogenesis and metabolism, resulting in increased energy expenditure, according to studies. When combined with the ketogenic effects of the gummies, this ingredient can aid in creating a more efficient fat-burning environment in the body.  A more comprehensive approach to weight management is the outcome. Citrus Sinensis (L.  Osbeck)</span></p> <ul> <li> <h4><strong> </strong><strong>Citrus sinensis</strong></h4> </li> </ul> <p><span style="font-weight: 400;">Citrus sinensis, more commonly known as sweet orange, is included because it contains a lot of vitamin C and other beneficial substances. This ingredient is believed to help boost metabolism and enhance fat oxidation, making it a valuable addition to Supra Keto BHB + ACV Gummies Australia.  In addition, the gummies' energizing citrus flavor may motivate consumers to stick with their weight loss plan. Citrus sinensis's antioxidants are also beneficial to well-being as a whole. Fucoidan (from Laminaria Japonica)</span></p> <ul> <li> <h4><strong>Kudzu Flower Extract</strong></h4> </li> </ul> <p><span style="font-weight: 400;"> Kudzu flower extract is known for its potential to reduce alcohol cravings, making it a beneficial ingredient for those who struggle with unhealthy eating habits.  By aiding in appetite control and reducing cravings, kudzu flower can make it easier for users to adhere to a ketogenic diet.  Furthermore, this extract possesses antioxidant properties that support general health, which can be particularly important during weight loss.  Including kudzu flower extract in the gummies provides a multi-faceted approach to weight management.</span></p> <ul> <li> <h4><strong>Berberine</strong></h4> </li> </ul> <p><span style="font-weight: 400;"> Berberine is a plant compound with potent health benefits, known primarily for lowering blood sugar and improving insulin sensitivity.  This ingredient can be particularly useful for individuals trying to lose weight, as it helps regulate metabolism and reduces fat accumulation.  The inclusion of berberine in Supra Keto BHB + ACV Gummies aids in achieving and maintaining a healthy weight by supporting weight loss objectives and promoting overall metabolic health. Xylitol</span></p> <p><span style="font-weight: 400;"> Xylitol is a natural sugar alcohol used as a sweetener in various low-calorie products.  It offers a sweet taste without the associated calories of traditional sugars, making it an excellent choice for a weight loss supplement.  Xylitol is known for its dental health benefits, as it can help reduce cavity-causing bacteria in the mouth.  By incorporating xylitol into Supra Keto BHB + ACV Gummies Australia, users can enjoy a delicious treat that doesn’t compromise their health goals.</span></p> <h3><strong> Benefits of Supra Keto BHB + ACV Gummies</strong></h3> <p><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> are appealing to people who want to lose weight and improve their overall health because they provide a wide range of advantages. The unique ingredients in the formula support each benefit, working together to enhance the product’s effectiveness.</span></p> <ul> <li> <h4><span style="font-weight: 400;"> </span><strong>Promotes Weight Loss</strong></h4> </li> </ul> <p><span style="font-weight: 400;"> One of the primary benefits of Supra Keto BHB + ACV Gummies is their ability to promote weight loss by helping the body enter ketosis.  In this metabolic state, the body becomes highly efficient at burning fat for energy instead of relying on carbohydrates.  This change makes it possible to lose weight more quickly. Many users report seeing a noticeable decrease in weight within just a few weeks of using the gummies, making them an appealing option for those seeking to shed excess pounds.</span></p> <p><span style="font-weight: 400;"> By encouraging the body to metabolize stored fat, Supra Keto BHB + ACV Gummies help minimize fat accumulation while improving energy levels.  People who cut back on carbohydrates may feel less hungry and less hungry for food, which helps them lose weight. The combination of natural ingredients not only enhances fat burning but also helps to stabilize blood sugar levels, reducing the likelihood of emotional eating and enabling users to stay on track with their dietary goals.</span></p> <ul> <li> <h4><strong> Boosts Metabolism</strong></h4> </li> </ul> <p><span style="font-weight: 400;">Another significant advantage of </span><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> is their ability to boost metabolism.  This effect is crucial for those trying to lose weight as a higher metabolic rate means more calories are burned throughout the day, even at rest.  Ingredients like fucoxanthin and berberine are essential in increasing metabolic activity, ensuring that users can maximize their weight loss potential.</span></p> <p> </p> <p style="text-align: center;"><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/ "><img src="https://i.ibb.co/p6g9jsbs/Supra-Keto-BHB-ACV-Gummies-2.webp" alt="Supra-Keto-BHB-ACV-Gummies-2" border="0" /></a>></p> <p> </p> <h1 style="text-align: center;"><span style="background-color: #ffcc99; color: #993300;"><em><a style="background-color: #ffcc99; color: #993300;" href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong>👉👀</strong><strong>𝙏𝙧𝙚𝙣𝙙𝙞𝙣𝙜</strong> <strong>𝙊𝙛𝙛𝙚𝙧</strong> <strong>𝙃𝙪𝙧𝙧𝙮</strong> <strong>𝙐𝙥</strong> <strong>𝙊𝙧𝙙𝙚𝙧</strong> <strong>𝙉𝙤𝙬</strong><strong>👀👈</strong></a></em></span></h1> <p><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> support metabolic processes, allowing users to maintain their energy levels while losing weight. Additionally, improved digestive health, which is essential to overall health, may result from a more effective metabolism. People will likely feel lighter and healthier when the body processes food and eliminates waste efficiently. As such, these gummies serve as a comprehensive solution for those on their weight loss journey.</span></p> <ul> <li> <h4><strong> Enhances Energy Levels</strong></h4> </li> </ul> <p><span style="font-weight: 400;"> Consuming Supra Keto BHB + ACV Gummies can provide a substantial boost in energy levels.  Users frequently experience decreased fatigue and increased stamina when the body switches to burning fat for fuel. This is especially beneficial for those embarking on a weight loss regimen, as maintaining high energy levels can encourage regular physical activity and make workouts more enjoyable.</span></p> <p><span style="font-weight: 400;"> The BHB and other ingredients in the gummies work to elevate energy levels by providing a quick and efficient energy source.  As users embrace their keto journey, they may feel more focused, alert, and ready to tackle their daily tasks.  This enhanced energy supports their fitness goals and improves overall productivity and mood throughout the day.</span></p> <ul> <li> <h4><strong>Provides Mental Focus</strong></h4> </li> </ul> <p><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> users may notice a significant improvement in their mental focus, which is important because cognitive health is just as important as physical health. Ingredients like BHB and saffron enhance alertness, concentration, and mood stabilization.  This increase in mental clarity and focus can be particularly beneficial for those navigating the challenges of dieting and weight loss.</span></p> <p><span style="font-weight: 400;"> People can better stick to their weight loss plans, make better food choices, and participate more fully in the activities they enjoy with better focus. A more balanced approach to health and wellness is the result of these mental benefits taken together, allowing users to succeed in their weight loss efforts. Are Supra Keto BHB + ACV Gummies associated with any side effects? As with any dietary supplement, users may wonder whether any side effects are associated with</span> <span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;">.  Generally, these gummies are made from natural ingredients, which helps minimize the likelihood of adverse effects.  However, individual reactions can vary based on personal health conditions and sensitivities.</span></p> <p><span style="font-weight: 400;">Some individuals might experience mild gastrointestinal discomfort, especially when starting a ketogenic diet or if they are sensitive to certain ingredients, like apple cider vinegar.  Common symptoms could include bloating, gas, or an upset stomach.  As the body adjusts to the ketogenic diet and ingredients, these effects may lessen over time. Anyone considering using</span> <span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> should consult their healthcare provider, particularly if they have pre-existing health conditions or are pregnant or nursing.  Doing so helps ensure the product aligns with their health needs and poses no risks.</span></p> <p><span style="font-weight: 400;">In summary, while side effects can occur, many users report a positive experience without significant issues.  Any potential discomfort can be reduced by drinking enough water and gradually introducing the gummies into your routine. Who makes Supra Keto BHB + ACV Gummies Australia?</span></p> <p><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> are produced by a reputable company specializing in high-quality dietary supplements.  This company has established a strong presence in the health and wellness industry, focusing on creating effective products that support weight loss and overall health.</span></p> <p><span style="font-weight: 400;">The formulation relies on well-researched ingredients known for their effectiveness in promoting ketosis and fat burning.  Additionally, the company’s commitment to quality and transparency helps dispel the notion of a scam.  They provide detailed information about their product, including ingredient sourcing, manufacturing processes, and customer support.</span></p> <p><span style="font-weight: 400;"> Furthermore, many users report positive experiences and noticeable weight loss results after incorporating Supra Keto BHB + ACV Gummies into their routines.  Despite individual variations in outcomes, there is general agreement that the product is beneficial and dependable rather than a con.</span></p> <p><span style="font-weight: 400;">A common question regarding dietary supplements is whether they are FDA approved.  It is essential to comprehend that dietary supplements do not receive the same level of FDA approval as pharmaceuticals. Instead, the regulation of dietary supplements falls under the Dietary Supplement Health and Education Act (DSHEA), which means that manufacturers are responsible for ensuring the safety and efficacy of their products before they reach the market.</span></p> <p><span style="font-weight: 400;">While</span> <span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> may not be FDA approved, they are manufactured in facilities adhering to strict safety and quality standards.  Good Manufacturing Practices (GMP) are commonly followed by reputable businesses to guarantee that their products are safe and effective for consumer use. Additionally, many manufacturers conduct third-party testing to validate the quality of their ingredients and formulations.</span></p> <p><span style="font-weight: 400;"> If you are considering using Supra Keto BHB + ACV Gummies US, you can feel confident that while they may not carry FDA approval, they likely meet stringent quality and safety standards.  To ensure you obtain a high-quality product, it’s always best to purchase from the company’s official website and check for transparency regarding their manufacturing processes.</span></p> <h3><strong> Where to buy Supra Keto BHB + ACV Gummies United States?</strong></h3> <p><span style="font-weight: 400;">Purchasing Supra Keto BHB + ACV Gummies is straightforward and can be done through the official website.  Since this is the only authorized place to purchase the item, you can rest assured that you will receive a genuine item that lives up to your expectations for quality and integrity. You can also take advantage of any promotions or discounts that may be offered if you buy directly from the official website. In addition, the website frequently provides in-depth information about the gummies, including their ingredients, advantages, and applications. For the best experience and to ensure your health and safety, avoid purchasing through third-party sellers or unauthorized platforms.  Doing so could result in receiving counterfeit or contaminated products.  To obtain your supply of Supra Keto BHB + ACV Gummies and take advantage of their benefits, please visit the official website. Conclusion</span></p> <p><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span><span style="font-weight: 400;"> present a compelling option for anyone seeking to enhance their weight loss journey and overall health.  With a carefully curated blend of natural ingredients, these gummies support the body’s transition into ketosis, promote fat burning, and provide various other health benefits.</span></p> <p><span style="color: #993366;"><a style="color: #993366;" href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>Supra Keto BHB + ACV Gummies</strong></a></span> <span style="font-weight: 400;">United States ease of use is one of their best qualities. Unlike traditional pills and supplements, these tasty gummies are easy to incorporate into daily routines, making weight management less chore.  Positive experiences have been shared by a lot of users, highlighting successful weight loss as well as improvements in energy levels, mental focus, and digestive health. These gummies, when combined with a healthy diet and regular exercise, can produce impressive results if realistic expectations are maintained. With various pricing options that provide significant savings for bulk purchases, Supra Keto BHB + ACV Gummies offer a cost-effective way to support your fitness and wellness goals.</span></p> <p><span style="font-weight: 400;"> Consult a medical professional before beginning any new supplement, especially if you have any existing health issues. Overall, Supra Keto BHB + ACV Gummies have the potential to be a useful addition to your weight loss plan, assisting you in achieving your objectives while providing you with a pleasant experience along the way.</span></p> <p><span style="font-weight: 400;"> </span></p> <p> </p> <p style="text-align: center;"><a href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/ "><img src="https://i.ibb.co/DDp6pW2P/Supra-Keto-BHB-ACV-Gummies.jpg" alt="Supra-Keto-BHB-ACV-Gummies" width="472" height="314" border="0" /></a></p> <p> </p> <h1 style="text-align: center;"><span style="background-color: #ffcc99; color: #666699;"><em><a style="background-color: #ffcc99; color: #666699;" href="https://www.knownwalk.com/2MCHKNWH/ZB4CLRD/"><strong>👉(Discount 50%) Click Here To Buy Supra Keto BHB + ACV Gummies For Special Discount👈</strong></a></em></span></h1> <h3><strong>Affiliate Disclosure:</strong></h3> <p><span style="font-weight: 400;"> The links contained in this product review may result in a small commission if you opt to purchase the product recommended at no additional cost to you.  This goes towards supporting our research and editorial team and please know we only recommend high quality products.</span></p> <h3><strong>Disclaimer:</strong></h3> <p><span style="font-weight: 400;"> Please understand that any advice or guidelines revealed here are not even remotely a substitute for sound medical advice from a licensed healthcare provider.  Make sure to consult with a professional physician before making any purchasing decision if you use medications or have concerns following the review details shared above.  Individual results may vary as the statements made regarding these products have not been evaluated by the Food and Drug Administration.  The efficacy of these products has not been confirmed by FDA-approved research.  These products are not intended to diagnose, treat, cure or prevent any disease.</span></p> <p><strong> </strong></p> <h3><span style="color: #808080;"><strong>RELATED TAGS :</strong></span></h3> <p><a href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>#SupraKetoBHB+ACVGummies</strong></a></p> <p><a href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>#SupraKetoBHB+ACVGummiesUS</strong></a></p> <p><a href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>#SupraKetoBHB+ACVGummiesUnitedStates</strong></a></p> <p><a href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>#SupraKetoBHB+ACVGummiesOfficial</strong></a></p> <p><a href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>#SupraKetoBHB+ACVGummiesReviews</strong></a></p> <p><a href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>#SupraKetoBHB+ACVGummiesBenefits</strong></a></p> <p><a href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>#SupraKetoBHB+ACVGummiesOffer</strong></a></p> <p><a href="https://www.facebook.com/SupraKETOBHBACVGummiesUS/"><strong>#SupraKetoBHB+ACVGummiesPrice</strong></a></p>
masani/SFT_cumulative_parity_length_16_bitwidth_1_512_512_Qwen2-1.5B_epoch_14_global_step_28	masani	2025-05-02T10:22:21Z	0	0	transformers	[ "transformers", "safetensors", "qwen2", "text-generation", "conversational", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T10:21:20Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
ninja/themes	ninja	2025-05-02T10:21:21Z	0	0	transformers	[ "transformers", "safetensors", "distilbert", "text-classification", "arxiv:1910.09700", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-classification	2025-05-02T10:21:01Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
thanhtantran/Vi-Qwen2-7B-RAG-rk3588-1.2.0	thanhtantran	2025-05-02T10:20:04Z	0	0	transformers	[ "transformers", "safetensors", "qwen2", "text-generation", "retrieval-augmented-generation", "text-generation-inference", "conversational", "vi", "base_model:Qwen/Qwen2-7B-Instruct", "base_model:finetune:Qwen/Qwen2-7B-Instruct", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T09:22:56Z	--- base_model: Qwen/Qwen2-7B-Instruct language: - vi library_name: transformers license: apache-2.0 pipeline_tag: text-generation tags: - retrieval-augmented-generation - text-generation-inference --- # Vi-Qwen2-7B-RAG-RK3588-1.2.0 This version of Vi-Qwen2-7B-RAG has been converted to run on the RK3588 NPU using w8a8 quantization. This model has been optimized with the following LoRA: Compatible with RKLLM version: 1.2.0 ## Useful links: [Official RKLLM GitHub](https://github.com/airockchip/rknn-llm) [RockhipNPU Reddit](https://reddit.com/r/RockchipNPU) [EZRKNN-LLM](https://github.com/Pelochus/ezrknn-llm/) Pretty much anything by these folks: [marty1885](https://github.com/marty1885) and [happyme531](https://huggingface.co/happyme531) Converted using https://github.com/c0zaut/ez-er-rkllm-toolkit # Original Model Card for base model, Vi-Qwen2-7B-RAG, below: ## Model Card: Vi-Qwen2-7B-RAG Mô tả mô hình: Vi-Qwen2-7B-RAG là một mô hình ngôn ngữ lớn được tinh chỉnh từ mô hình cơ sở Qwen2-7B-Instruct (https://huggingface.co/Qwen/Qwen2-7B-Instruct) phục vụ cho RAG tasks. Mô hình được đào tạo trên tập dữ liệu tiếng Việt với mục tiêu cải thiện khả năng xử lý ngôn ngữ tiếng Việt và nâng cao hiệu suất cho các tác vụ Retrieval Augmented Generation (RAG). Mục đích sử dụng: Mô hình Vi-Qwen2-7B-RAG được thiết kế chuyên biệt cho RAG (ngữ cảnh chấp nhận lên đến 8192 tokens), vì vậy nó có thể giải quyết các trường hợp sau: * Khả năng chống nhiều: Mô hình trích xuất thông tin hữu ích từ các tài liệu nhiễu. ( 1 positive + 4 negative hoặc 1 negative) * Loại bỏ negative: Mô hình từ chối trả lời câu hỏi khi kiến thức cần thiết không có trong bất kỳ tài liệu nào được truy xuất. (1-6 negative) * Tích hợp thông tin: Mô hình trả lời các câu hỏi phức tạp đòi hỏi phải tích hợp thông tin từ nhiều tài liệu. ( 2 part positive + 3 negative hoặc 3 part positive + 2 negative) * Xác đinh positive/negative: Mô hình xác định xem ngữ cảnh có chứa câu trả lời cho câu hỏi hay không. (độ chính xác xấp xỉ 99%) Ngoài ra, chúng tôi cũng triển khai các mô hình nhỏ hơn phù hợp với mục đích sử dụng khác nhau như Vi-Qwen2-1.5B-RAG (https://huggingface.co/AITeamVN/Vi-Qwen2-1.5B-RAG) và Vi-Qwen2.5-3B-RAG (https://huggingface.co/AITeamVN/Vi-Qwen2-3B-RAG) * Ngoài RAG task, bạn vẫn có thể chatbot bình thường với model của chúng tôi. Thậm chí có thể hỏi các câu hỏi liên tục với ngữ cảnh đầu vào. Hạn chế: Vì mô hình chỉ được thiết kế chuyên biệt cho RAG task, nên có thể gặp 1 số hạn chế sau: * Không đảm bảo độ chính xác về các câu hỏi liên quan đến chính trị, xã hội,... * Có thể thể hiện thành kiến hoặc quan điểm không phù hợp. Benchmarks Chúng tôi đã đánh giá một số mô hình LLMs cho RAG task sử dụng tập dữ liệu được tạo thủ công bởi chúng tôi [EvalRAGData](https://huggingface.co/datasets/AITeamVN/EvalRAGData). Đánh giá được thực hiện bởi con người chấm điểm trên thang điểm 10. Dưới đây là kết quả chi tiết: ![image/png](https://cdn-uploads.huggingface.co/production/uploads/66e159e8a533df3d827ec23f/p-JMK67C0BqIAi723YA1p.png) Ngoài ra, chúng tôi cũng tiến hành benchmark trên VMLU leaderboard: \| Model \| Stem \| Social Science \| Humanities \| Others \| Average \| \|----------------------\|------------\|------------\|------------\|-------------\|--------------\| \| Vi-Qwen2-7B-RAG \| 60.22 \| 57.8 \| 52.67 \| 51.08 \| 56.04 \| Các cách sử dụng: #### 1. Sử dụng cơ bản Ngữ cảnh đầu vào chỉ chứa 1 ngữ cảnh (1 postive hoặc 1 negative). ```python import torch from transformers import AutoModelForCausalLM, AutoTokenizer path = 'AITeamVN/Vi-Qwen2-7B-RAG' model = AutoModelForCausalLM.from_pretrained( path, torch_dtype=torch.bfloat16, device_map="auto", use_cache=True ) tokenizer = AutoTokenizer.from_pretrained(path) system_prompt = "Bạn là một trợ lí Tiếng Việt nhiệt tình và trung thực. Hãy luôn trả lời một cách hữu ích nhất có thể." template = '''Chú ý các yêu cầu sau: - Câu trả lời phải chính xác và đầy đủ nếu ngữ cảnh có câu trả lời. - Chỉ sử dụng các thông tin có trong ngữ cảnh được cung cấp. - Chỉ cần từ chối trả lời và không suy luận gì thêm nếu ngữ cảnh không có câu trả lời. Hãy trả lời câu hỏi dựa trên ngữ cảnh: ### Ngữ cảnh : {context} ### Câu hỏi : {question} ### Trả lời :''' # Ví dụ context = '''Thuốc Insuact 10 trị bệnh gì? Thuốc Insuact 10mg có thành phần chính là Atorvastatin. Thuốc Insuact 10 có tác dụng làm giảm cholesterol và triglycerid trong máu ở bệnh nhân tăng cholesterol máu nguyên phát, rối loạn lipid máu hỗn hợp. 1. Thuốc Insuact 10 trị bệnh gì? Thuốc Insuact 10 thuộc nhóm thuốc điều trị rối loạn lipid máu, có thành phần chính là Atorvastatin 10mg. Atorvastatin có tác dụng làm giảm cholesterol, ức chế enzym tạo cholesterol ở gan. Atorvastatin làm giảm cholesterol chung bao gồm cholesterol LDL , triglycerid trong máu. Thuốc Insuact 10mg được bào chế dưới dạng viên nén bao phim, được chỉ định dùng trong những trường hợp sau: Ðiều trị hỗ trợ tăng cholesterol máu nguyên phát và rối loạn lipid máu hỗn hợp trên bệnh nhân đang áp dụng chế độ ăn kiêng để làm giảm cholesterol toàn phần , cholesterol LDL , apolipoprotein B, triglycerid và tăng cholesterol HDL . Insuact 10 cũng được dùng để điều trị rối loạn betalipoprotein trong máu nguyên phát. Ðiều trị hỗ trợ tăng cholesterol trong máu có tính gia đình đồng hợp tử trên bệnh nhân đang áp dụng các biện pháp làm giảm lipid khác để làm giảm cholesterol toàn phần và cholesterol LDL. 2. Liều dùng và cách dùng thuốc Insuact 10 Cách dùng thuốc Insuact 10: Thuốc được dùng theo đường uống, uống khi bụng đói hoặc no đều được, có thể uống vào bất cứ lúc nào trong ngày. Liều dùng thuốc Insuact 10mg khởi đầu là 10mg/lần/ngày, tối đa là 80mg/lần/ngày. Liều dùng thuốc Insuact 10 tùy vào mục đích điều trị cụ thể như sau: Tăng cholesterol máu nguyên phát và rối loạn lipid máu phối hợp: 10mg/lần/ngày, sau 2 - 4 tuần sẽ thấy hiệu quả của thuốc. Thuốc cần được được sử dụng duy trì trong thời gian dài để có hiệu quả. Tăng cholesterol trong máu có tính gia đình đồng hợp tử: Liều thường dùng là thuốc Insuact 10mg /lần/ngày và tối đa là 80mg/lần/ngày. Rối loạn lipid máu nghiêm trọng ở trẻ từ 10 - 17 tuổi: 10mg/lần/ngày, sau đó tăng lên 20mg/lần/ngày tùy vào cơ địa, tiến triển bệnh và khả năng dung nạp thuốc của người bệnh. Thời gian điều chỉnh liều thuốc tối thiểu là 4 tuần. 3. Tác dụng phụ của thuốc Insuact 10mg Thuốc Insuact 10 có thể gây một số tác dụng phụ không mong muốn với tần suất như sau: Thường gặp: Viêm mũi - họng, phản ứng dị ứng, tăng đường huyết, nhức đầu, đau thanh quản, chảy máu cam , đau cơ, co thắt cơ, đau khớp, sưng khớp, đau các chi, đau lưng, xét nghiệm gan bất thường, tăng creatine kinase trong máu, buồn nôn, khó tiêu, đầy hơi, táo bón, tiêu chảy. Ít gặp: Insuact 10 ít gây hạ đường huyết, tăng cân, chán ăn, mất ngủ, gặp ác mộng, choáng váng, dị cảm, mất trí nhớ, giảm cảm giác, loạn vị giác , nôn, đau bụng, ợ hơi, viêm tụy, viêm gan, nổi mày đay , phát ban, ngứa, rụng tóc, đau cổ, mỏi cơ, mệt mỏi, suy nhược, đau ngực, phù ngoại biên, sốt, xuất hiện bạch cầu trong nước tiểu, nhìn mờ, ù tai. Hiếm gặp: Insuact 10 hiếm khi làm giảm tiểu cầu, bệnh lý thần kinh ngoại biên, hoa mắt, ứ mật, phù thần kinh, nổi hồng ban, hội chứng hoại tử da nhiễm độc , hội chứng Stevens-Johnson , bệnh cơ, viêm cơ, tiêu cơ vân, bệnh gân, đôi khi nghiêm trọng hơn có thể đứt gân. Rất hiếm gặp: Insuact 10 rất hiếm khi gây sốc phản vệ , mất thính giác , suy gan , hội chứng to vú ở nam giới. Không rõ tần suất: Hoại tử cơ tự miễn trung gian. 4. Một số lưu ý khi dùng thuốc Insuact 10mg Không dùng thuốc Insuact với người bị quá mẫn với thành phần của thuốc, người có bệnh gan hoạt động hoặc tăng transaminase huyết thanh vô căn kéo dài, phụ nữ đang mang thai hoặc nuôi con cho bú, phụ nữ đang có ý định mang thai. Thuốc Insuact 10mg chỉ được dùng ở bệnh nhân có nguy cơ xơ vữa mạch máu cao do tăng cholesterol trong máu và phải kết hợp với chế độ ăn kiêng ít chất béo bão hòa , ít cholesterol và người bệnh đang áp dụng các biện pháp điều trị không dùng thuốc khác. Trước khi điều trị với Insuact 10 , người bệnh cần được loại trừ các nguyên nhân thứ phát gây tăng cholesterol bao gồm suy tuyến giáp , tiểu đường khó kiểm soát, hội chứng thận hư, nghiện rượu, bệnh gan tắc nghẽn, rối loạn protein trong máu, .... Ngoài ra, người bệnh cũng cần được kiểm tra, đo lường nồng độ lipid máu. Trước khi điều trị với thuốc Insuact 10mg , cần kiểm tra chức năng gan và trong quá trình dùng thuốc, người bệnh cần theo dõi, kiểm tra chức năng gan thường xuyên. Người có tiền sử mắc bệnh gan do rượu, bị nghiện rượu cần thận trọng khi dùng Insuact 10 . Trước khi dùng thuốc, người bệnh cần được cảnh báo nguy cơ gặp phải các vấn đề về cơ như căng cơ , đau cơ, yếu cơ. Thuốc Insuact 10mg có thể tương tác với các thuốc: Tăng nguy cơ tổn thương cơ (đặc biệt cơ vân) khi dùng đồng thời với thuốc điều trị viêm gan C và HIV, hạ cholesterol máu nhóm fibrat khác, thuốc ức chế miễn dịch; tăng thời gian chảy máu hoặc thời gian prothrombin khi dùng phối hợp với thuốc chống đông, indandione; tăng nhẹ nồng độ digoxin trong huyết thanh khi dùng phối hợp với Digoxin''' question = '''Insuact 10mg được chỉ định dùng trong những trường hợp nào và liều dùng cụ thể ra sao?''' conversation = [{"role": "system", "content": system_prompt }] conversation.append({"role": "user", "content": template.format(context = context, question = question)}) text = tokenizer.apply_chat_template( conversation, tokenize=False, add_generation_prompt=True) model_inputs = tokenizer(text,return_tensors="pt").to(model.device) generated_ids = model.generate( model_inputs.input_ids, max_new_tokens=2048, temperature = 0.1, #top_p=0.95, #top_k=40, ) generated_ids = [ output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids) ] response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] print(response) '''Trả lời: Dựa trên ngữ cảnh cung cấp, Insuact 10mg được chỉ định dùng trong các trường hợp sau: 1. Điều trị hỗ trợ tăng cholesterol máu nguyên phát và rối loạn lipid máu hỗn hợp: - Liều dùng: 10mg/lần/ngày. - Hiệu quả: Sau 2 - 4 tuần sử dụng, bệnh nhân sẽ thấy hiệu quả của thuốc. - Lưu ý: Thuốc cần được sử dụng duy trì trong thời gian dài để có hiệu quả. 2. Điều trị hỗ trợ tăng cholesterol trong máu có tính gia đình đồng hợp tử: - Liều dùng: 10mg/lần/ngày (liều thường dùng) và tối đa 80mg/lần/ngày. - Lưu ý: Liều lượng có thể được điều chỉnh tùy theo cơ địa, tiến triển bệnh và khả năng dung nạp thuốc của người bệnh. 3. Rối loạn lipid máu nghiêm trọng ở trẻ từ 10 - 17 tuổi: - Liều dùng: 10mg/lần/ngày, sau đó có thể tăng lên 20mg/lần/ngày tùy theo tình trạng bệnh. - Thời gian điều chỉnh liều: Thời gian điều chỉnh liều tối thiểu là 4 tuần. Lưu ý chung: - Thuốc được dùng theo đường uống, uống khi bụng đói hoặc no đều được, có thể uống vào bất cứ lúc nào trong ngày. - Không dùng thuốc Insuact 10mg với người bị quá mẫn với thành phần của thuốc, người có bệnh gan hoạt động hoặc tăng transaminase huyết thanh vô căn kéo dài, phụ nữ đang mang thai hoặc nuôi con cho bú, phụ nữ đang có ý định mang thai. - Cần kiểm tra chức năng gan trước và trong quá trình điều trị. ''' ``` #### 2. Khả năng chống nhiều ```python # ví dụ context = '''- Ngữ cảnh 1: . Và dù nguyên nhân thế nào, thì việc đánh người khác đều không thể chấp nhận, không ai có quyền xâm phạm thân thể của người khác, nhất là đánh những đứa trẻ là bạn của con cái mình. Lựa chọn kể với cha mẹ, người lớn về các mâu thuẫn học đường, là quyết định rất đúng của các em. Nhưng người lớn, đặc biệt những ông bố, bà mẹ cần ứng xử ra sao trước xung đột này của các con, thưa bà? Đứng ngoài mâu thuẫn bằng sự khách quan và trách nhiệm nhất có thể. Điều này giúp chúng ta đủ bình tĩnh để làm việc với tất cả các bên liên quan, từ giáo viên, bạn của con, ban giám hiệu để tìm hiểu câu chuyện và tìm kiếm cách giải quyết tích cực, trên cơ sở phối hợp nhà trường. Người lớn không thể chỉ nghe một tai và đặc biệt không nên tự xử. Phụ huynh, kể cả học sinh tự xử các vấn đề học đường là điều rất nguy hiểm và cho thấy sự coi thường pháp luật . Vụ việc ở Tuyên Quang vừa rồi là ví dụ. Các em hoàn toàn có thể phản ứng bằng cách trình bày, gửi yêu cầu, kiến nghị lên nhà trường, nhờ phụ huynh làm việc với ban giám hiệu để có biện pháp giải quyết nếu cô giáo sai, không nên đồng loạt dồn cô giáo vào tường một cách bạo lực và trái đạo đức, tôn ti trật tự như vậy. Ngoài ra, chúng ta cũng có rất nhiều cơ quan chức năng bảo vệ phụ huynh và con em, với những quyền về khiếu nại, tố cáo. Chúng ta nói nhiều về trường học an toàn. Trong những câu chuyện học sinh bị hành hung thế này, có lẽ cũng cần làm rõ vai trò, trách nhiệm của nhà trường? TPHCM và nhiều địa phương đang xây dựng môi trường trường học hạnh phúc, tiêu chí là yêu thương, an toàn, tôn trọng. Không chỉ phòng ngừa, nhà trường còn phải tích cực vào cuộc xử lý các mâu thuẫn học đường, hạn chế tối đa nguy cơ mất an toàn cho học sinh, giáo viên. Đặc biệt, giải quyết câu chuyện bạo lực học đường phải triệt để, tuyệt đối không nửa vời vì nửa vời sẽ tiềm ẩn nguy cơ rất lớn dẫn đến những vụ việc tương tự, với mức độ nghiêm trọng hơn. Vụ việc em M. ở Nha Trang tự vẫn với lá thư tuyệt mệnh bị đổ oan đầu tháng 10 vừa qua là một ví dụ về giải quyết không triệt để. Việc xây dựng trường học hạnh phúc nếu triển khai “đến nơi đến chốn”, sẽ góp phần rất lớn cải thiện tình trạng bạo lực học đường, tạo môi trường sống và học tập bình an cho các con. Từ nhiều sự vụ học sinh bạo hành lẫn nhau, giáo viên bạo hành học sinh, phụ huynh hành hung giáo viên và bạn của con. Tam giác phối hợp bảo vệ học sinh là nhà trường - gia đình - xã hội phải chăng đang có một lỗ hổng lớn, thưa bà? Câu chuyện này có liên quan đến niềm tin của phụ huynh với nhà trường. Tại sao phụ huynh lại chọn cách tự xử? Chúng ta cũng cần phải xem lại cách giải quyết vấn đề của nhà trường đã rốt ráo chưa, coi trọng lợi ích của tất cả các bên liên quan chưa hay chỉ đang xoa dịu? Người ta chỉ tìm đến nhau khi có niềm tin vào nhau. Thực trạng phụ huynh chọn cách chuyển trường cho con cũng nói lên điều này. Đây là cách chạy trốn của phụ huynh với mong muốn con được an toàn, hạnh phúc hơn ở môi trường mới. Xây dựng niềm tin cho phụ huynh, xã hội cần được chú trọng và với mỗi một trường hợp phụ huynh yêu cầu chuyển trường cho con - đang rất phổ biến - nhà trường cần xét kỹ các nguyên nhân và hóa giải. Xin bà cho biết đâu là giải pháp căn cơ cho tất cả những câu chuyện bạo lực nói trên? Để trẻ không là nạn nhân của bạo lực học đường, phụ huynh cần đồng hành và giúp con có sự hiểu biết, ý thức trước vấn đề này. Dạy con kỹ năng giao tiếp, quản lý cảm xúc rất quan trọng và điều này không thể chỉ dựa vào những khóa học kỹ năng sống, mà là từ cách cư xử của người lớn, cha mẹ, thầy cô. Không có tấm gương nào tốt hơn cho con trẻ bằng ứng xử, hành vi của người lớn. Vì vậy, không thể đòi hỏi trẻ nói không với bạo lực học đường khi trong chính từng gia đình, xã hội, người lớn vẫn đối xử với nhau bằng bạo lực. - Ngữ cảnh 2: Tại sao triều Thanh có rất nhiều thân vương nhưng chẳng có ai dám tạo phản? Không giống như những triều đại trước đó, triều Thanh dù có sự tranh giành ngai vàng khốc liệt giữa các hoàng tử nhưng lại chẳng bao giờ xảy ra thế cục các thân vương tạo phản. Chính vì 3 lý do lớn này đã khiến cho triều đại nhà Thanh khác hẳn triều đại nhà Đường và nhà Minh. Trong thời cổ đại, các vương công quý tộc để tranh giành vương vị của mình, giữa huynh đệ ruột thịt với nhau dường như đều xảy ra đấu đá, hãm hại lẫn nhau, coi nhau như kẻ thù không đội trời chung, có ta thì không có ngươi, có ngươi thì sẽ chẳng có ta, điều này hoàn toàn không phải là điều gì xa lạ. Vậy thì tại sao ngai vàng lại có sức hút lớn đến thế? Không chỉ là đàn ông khát khao quyền lực, mà quan trọng hơn là hoàng đế có thể có được hậu cung rộng lớn, trong hậu cung còn có vô số các mỹ nữ quốc sắc thiên hương. Nhiều phi tần như vậy, đương nhiên hoàng đế cũng sẽ có rất nhiều con cái, không tính đến con gái, chỉ riêng những vị hoàng tử, để có thể có được hoàng vị, họ tranh giành nhau bằng cả sinh mạng. Vậy thì ai là người được lựa chọn để thừa kế ngai vàng, ai mới có thể gánh được trọng trách trị vì đất nước? Đa phần đều theo tục lệ truyền cho con trai đích tôn (con trai do hoàng hậu sinh ra) hoặc con trai trưởng (con trai đầu tiên của hoàng đế). Cho dù tục lệ này có lịch sử lâu đời nhưng nó cũng có những khuyết điểm rất lớn, đó chính là nếu như năng lực và chí hướng của con trai đích tôn hoặc con trai trưởng không thể gánh vác được ngai vị, nếu để anh ta lên ngôi hoàng đế, vậy thì đất nước sẽ rơi vào cục diện suy vong. Còn có một khuyết điểm nữa đó chính là những người con trai có dã tâm lớn khác sẽ không phục việc con trai đích hoặc con trai trưởng kế thừa ngôi báu, họ sẽ khởi binh tạo phản cũng là chuyện rất dễ xảy ra. Ví dụ như trong thời Đường của Trung Quốc, Đường Cao Tổ Lý Uyên đem binh tiêu diệt nhà Tùy thối nát, đồng thời lập nên nhà Đường, vốn dĩ ông cũng dựa theo tục lệ lập con trai trưởng là Lý Kiến Thành làm Thái tử nhưng con trai thứ là Lý Thế Dân lại không phục với sự sắp xếp này. Vì năng lực của ông xuất chúng, văn võ song toàn, còn lập được không ít công lao to lớn trong cuộc chiến tranh tiêu diệt nhà Tùy cùng cha mình, đương nhiên không chịu thấp hơn kẻ khác một bậc. Thế nên đã phát động binh biến Huyền Vũ Môn, trong cuộc binh biến tạo phản này, đích thân ông đã giết chết huynh trưởng của mình, đồng thời ép cha mình là Lý Uyên phải truyền ngôi cho mình. Hay như trong thời nhà Minh của Trung Quốc, trước khi Chu Nguyên Chương chọn người lập làm Thái tử, con trai trưởng Chu Tiêu đã qua đời vì bệnh nặng, thế nên Chu Nguyên Chương đã lập cháu đích tôn của mình làm Thái tử kế thừa vương vị, nhưng em trai của Chu Tiêu là Chu Đệ lại không phục lựa chọn này của Chu Nguyên Chương. Theo lý mà nói thì sau khi anh trai Chu Tiêu qua đời, ông đã có tư cách thừa kế ngai vàng nhưng Chu Nguyên Chương nhất quyết không chọn ông mà lại chọn người cách thế hệ để truyền ngôi. Điều này khiến Chu Đệ với thế lực to lớn không thể nuốt nổi cục tức này, vì thế Chu Tiêu vừa qua đời thì ông đã vội vã khởi binh tạo phản, giết chết cháu trai ruột của mình rồi tự xưng vương. Vậy thì tại sao trong triều Thanh có rất nhiều thân vương như vậy mà lại chẳng có ai đứng ra tạo phản? Đầu tiên phải nói về bối cảnh xã hội trong thời kỳ này. Triều Thanh từ khi thành lập, cũng giống với những triều đại khác, đều có rất nhiều thân vương. Nếu người dân bình thường muốn làm hoàng đế, vậy thì đó là điều hoàn toàn không thể, nhưng đối với những vương công quý tộc trong hoàng thất mà nói, họ đương nhiên sẽ có rất nhiều cơ hội, đặc biệt là những thân vương nắm đại quyền quân sự , họ chính là mối đe dọa lớn nhất đối với nhà vua. Vì thế, các đời hoàng đế đều sẽ nghĩ đủ mọi cách để áp chế, kiểm soát họ, tránh việc họ khởi binh tạo phản. Triều Thanh có lịch sử hơn 300 năm, cũng đã cho ra đời vô số thân vương, đặc biệt là cuối thời Thanh, khi Trung Quốc rơi vào cảnh khốn khó, sau khi Từ Hy Thái Hậu cầm quyền thì thế cục này càng được thể hiện rõ rệt hơn. Nhưng cho dù là một người phụ nữ cầm quyền thì cũng chẳng có một vị thân vương hoàng tộc nào đứng ra tạo phản. Có 3 nguyên nhân sau: Thứ nhất, thân vương triều Thanh không thể nối ngôi, nếu muốn tiếp tục duy trì danh phận thân vương, vậy thì bắt buộc phải có được sự đồng ý của hoàng đế và phải lập được công lao cho đất nước. Thứ hai, triều đình tiến hành giám sát nghiêm ngặt đối với các thân vương, họ không hề có cơ hội để tạo phản. Thứ ba, các thân vương không thể giao thiệp quá sâu với các đại thần, quan lại khác, điều này cũng khiến các thân vương rơi vào cảnh bị cô lập, thế nên càng không có cơ hội để cấu kết với người khác hòng tạo phản. - Video: Ngắm sự kỳ vĩ và lộng lấy của Tử Cấm Thành từ trên cao. Nguồn: Sky Eye. - Ngữ cảnh 3: . Cùng điều chỉnh với con là điều rất quan trọng bởi vì trẻ sẽ tự tin để tự đặt những giới hạn cho chính mình khi lớn lên”, TS Nguyễn Thị Thanh đưa ra lời khuyên. “Khi con mắc sai lầm, hãy giúp chúng tìm những cách khác tốt hơn. Đơn cử dùng hậu quả để dạy cho chúng bài học, điều đó tốt hơn rất nhiều việc xử phạt. Nếu cha mẹ chỉ biết trừng phạt, sẽ nhận được lời xin lỗi nhưng không thể giúp trẻ tỉnh ngộ. Bởi chúng chỉ biết được mình đã sai mà không biết sai ở chỗ nào và làm thế nào mới là đúng” - Ngữ cảnh 4: . “MẤT ĐI CHA MẸ Ở TUỔI ĐẸP NHẤT CỦA NGƯỜI PHỤ NỮ CÀNG KHIẾN TÔI PHẢI MẠNH MẼ” - Làm con của nghệ sĩ Thanh Hiền, Đinh Y Nhung cảm nhận sợi dây liên kết giữa hai mẹ con thế nào? Má Thanh Hiền là người rất tuyệt vời. Hai má con hồi xưa từng làm phim truyền hình với nhau rồi, cho nên khi tái hợp thì không mấy bỡ ngỡ. Khi đối diễn, hai má con rất ăn ý, như người thân ruột thịt vậy đó. - Khi thể hiện những phân cảnh cảm động trong phim, có khi nào chị thấy nhớ mẹ không? Có chứ, nhất là ở những phân đoạn gia đình sum họp, tự nhiên mình bị buồn. Ai cũng muốn có cha, có mẹ, ai cũng muốn Tết được chạy về bên gia đình. Trong mười mấy, hai chục năm qua, Nhung bị chạnh lòng. Tuy nhiên, chỉ trong tích tắc, tôi tự trấn an rằng, mình đang quay phim, đang hóa thân vào nhân vật nên không thể xao lãng được. Mình là con người mà, cũng có lúc tâm trạng vui buồn bất chợt, nhưng Nhung luôn cố gắng lấy lại phong độ liền. - Mất ba mẹ từ sớm, không có chỗ dựa tinh thần, cô gái trẻ Đinh Y Nhung năm đó có nhận những lời mời gọi khiếm nhã không? Trước đây, Nhung không có bạn bè nhiều, chủ yếu chỉ lo đi học, đi làm để lo cho cuộc sống thôi. Nên Nhung không phải đón nhận những lời mời gọi nào hết. - Mất mát từ quá khứ có ảnh hưởng gì đến suy nghĩ về tương lai của chị sau này, ví dụ khi có con thì sẽ bù đắp, chăm sóc cho con nhiều hơn? Năm ba mẹ mất thì mình vẫn còn khá trẻ, thật ra cái tuổi đó là tuổi đẹp của người phụ nữ. Sau đó, tôi đi làm, rồi yêu đương và lập gia đình. Có rất nhiều thứ hối tiếc để nói về Nhung của thời điểm đó. Thứ nhất là mình chưa thành công, thứ hai là mình chưa trả hiếu cho cha mẹ, thứ ba là mình còn bấp bênh. Nhung lúc đó lì lợm lắm, không cho phép mình ngã, bằng mọi giá phải tiến về trước dù có hàng ngàn chông gai ngăn cản. Có lúc tôi bị người này cười, người kia mỉa, nhưng mà mình vẫn cố bước đi. Người ta có cười thì cũng không mang lại cho mình được gì, tôi chỉ biết làm hết khả năng để lo cho bản thân, lo cho em của mình. Hiện, con cái Nhung đã đi nước ngoài rồi. Bé đang học đại học về âm nhạc, còn em mình cũng đã lớn rồi. Đối với Nhung ngay lúc này thì không phải thành công hay hoàn hảo lắm, nhưng ít nhất là tôi đã cố gắng để tự chịu trách nhiệm với cuộc đời mình. Mất cha, mất mẹ, đối với một người hai mươi mấy tuổi thì điều cần nhất lúc đó là có được gia đình ở bên. Nhưng mình không có chỗ dựa tinh thần thì càng phải mạnh mẽ hơn nữa. Tôi tự gặm nhấm nỗi đau mất người thân trong một thời gian dài, có khi đến cả bạn bè cũng không hề biết. Một thời gian sau, bạn bè thời và mọi người mới biết. Còn người hâm mộ, đồng nghiệp trong nghề gần như không biết chuyện ba mẹ Nhung mất sớm, chỉ có vài người chơi thân với nhau biết thôi. Sau này, dần dần tâm lý dần ổn định thì mình mới bắt đầu chia sẻ. “CON ĐI DU HỌC, TÔI DẶN BÉ CÁI GÌ KHÔNG TỐT THÌ MÌNH BỎ QUA” - Đinh Y Nhung từng tiết lộ mình rất thân với con gái. Có vẻ như quyết định để con đi du học là không hề dễ dàng? Thật sự là không có ba mẹ nào muốn con mình đi xa, nhưng việc du học lại là quyết định của bé. Con Nhung bày tỏ muốn học đại học ở nước ngoài và muốn đi sớm để thực hiện ước mơ. Nhưng lúc đó con còn nhỏ quá, phải đợi đến năm con 17 tuổi thì Nhung mới quyết định cho bạn nhỏ đi. Con cái từ nhỏ ở với bố mẹ giờ lại đi xa thì tất nhiên người làm cha làm mẹ cùng phải thấy sốc, thấy buồn. Nhưng Nhung hoàn toàn tôn trọng quyết định của con về việc chọn ngành nghề và tương lai của mình. Ba mẹ sẽ đứng sau và là người đưa cho con những lời khuyên và chỉ có thể đồng hành cùng con tới một mốc thời gian nào đó. Về sau, con phải đi làm và tự có trách nhiệm với cuộc đời của mình. - Có con gái đang ở tuổi lớn lại xa bố mẹ và tiếp xúc một nền văn hóa phương Tây cởi mở, Đinh Y Nhung đã khuyên dạy và đồng hành với con như thế nào? Ngay khi ở Việt Nam, con gái Nhung đã được theo học trường quốc tế. Hai mẹ con cũng có rất nhiều dịp để tâm sự và chia sẻ với nhau. Ngay từ nhỏ, Nhung đã cho bé được tiếp xúc song song giữa hai nền văn hóa để con không bỡ ngỡ. Mình là người Việt nên đương nhiên vẫn dạy con theo văn hóa Á Đông là chủ yếu. Nhung vẫn luôn tạo điều kiện để con cảm nhận những nét đẹp trong nền văn hóa quê hương. Văn hóa phương Tây thì xa lạ hơn nhưng Nhung cũng khuyên con rằng điều gì hay thì mình học hỏi, cái gì không tốt thì mình nên bỏ qua. Tất nhiên mình không thể theo sát con, nhất là khi bé đang ở độ tuổi mới lớn, có nhiều sự hiếu kỳ. Tuy nhiên, Nhung cũng không quá lo lắng vì qua quá trình học tập ở các trường quốc tế, bé cùng đã được làm quen dần với văn hóa phương Tây. Bé muốn làm bạn với mẹ nên có nhiều thứ bé muốn hỏi, muốn tiếp thu thì hai mẹ con lại ngồi xuống chia sẻ, tâm sự với nhau. Nhung tin, con luôn tỉnh táo để đưa ra những quyết định cho bản thân mình. Nhung không dám nói trước, nhưng hiện tại con vẫn luôn biết nói cảm ơn, xin phép trước khi làm bất cứ điều gì nên mình vẫn rất tin tưởng con. - Chị nhận xét thế nào về tính cách của con gái? Phải chăng bé là phiên bản nhí của chị? Con gái Nhung có nhiều nét giống mẹ. - Ngữ cảnh 5: . - Trong quá trình quay phim, má Hiền có tham gia đóng góp ý kiến cho dàn diễn viên trẻ không? Không phải má Hiền muốn làm thầy nhưng có những lúc cũng lên tiếng góp ý một vài điều. Những lúc như thế các bạn diễn viên trẻ cũng nghe theo. Phải giúp đỡ nhau như vậy mới có thể tạo ra được một tác phẩm hay. Chứ có nhiều diễn viên trẻ có lòng tự trọng cao, chỉ khi đạo diễn nói mới nghe thì không bao giờ có thể tiến bộ được. Má Hiền phải cảm thấy hạnh phúc mới đúng, càng ngày xung quanh càng có nhiều bạn bè, các mối quan hệ nhờ đó mà cũng ngày càng đông. Ngày xưa, khi tham gia là Mùi Ngò Gai chưa có ai biết nhưng sau khi phim phát sóng mọi người đã nhớ đến má. Hiện tại, má Hiền đã được đạo diễn, nhà sản xuất mời đóng phim đó đã là một niềm hạnh phúc. Đi ra đường hay đi khám bệnh, người ta cứ vây xung quanh quan tâm hỏi han khiến má cảm thấy rất vui lắm (cười). Cảm ơn nghệ sĩ Thanh Hiền đã nhận lời tham gia phỏng vấn với SAOstar. Chúc má có nhiều sức khỏe để tiếp tục làm nghề!''' question = '''Tại sao các thân vương triều Thanh không dám tạo phản mặc dù có nhiều quyền lực và cơ hội?''' '''Trả lời: Các thân vương triều Thanh không dám tạo phản mặc dù có nhiều quyền lực và cơ hội vì có ba lý do chính: 1. Không thể nối ngôi: Để duy trì danh phận thân vương, họ phải có sự đồng ý của hoàng đế và phải lập được công lao cho đất nước. Điều này hạn chế cơ hội của họ để trở thành hoàng đế. 2. Giám sát nghiêm ngặt: Triều đình tiến hành giám sát nghiêm ngặt đối với các thân vương, không cho họ cơ hội để tạo phản. 3. Bị cô lập: Các thân vương không thể giao thiệp quá sâu với các đại thần, quan lại khác, điều này khiến họ bị cô lập và không có cơ hội cấu kết với người khác để tạo phản.''' ``` #### 3. Loại bỏ negative ```python context = '''- Ngữ cảnh 1: Thái Bình hướng đến là trung tâm công nghiệp, năng lượng của vùng Với tiềm năng sẵn có, quy hoạch tỉnh Thái Bình thời kỳ 2021-2030, tầm nhìn đến năm 2050 xác định tỉnh sẽ phát triển công nghiệp theo hướng hiện đại, bền vững dựa trên nghiên cứu phát triển điện gió, điện khí, cân bằng lượng phát thải. Sáng 5/3, UBND tỉnh Thái Bình tổ chức Hội nghị công bố quy hoạch của tỉnh thời kỳ 2021-2030, tầm nhìn đến năm 2050 và xúc tiến đầu tư tỉnh Thái Bình. Phát biểu tại hội nghị, Phó Chủ tịch Thường trực UBND tỉnh Nguyễn Quang Hưng cho biết: Mục tiêu của quy hoạch là đến năm 2030, Thái Bình trở thành địa phương thuộc nhóm phát triển khá và là một trong những trung tâm phát triển công nghiệp của vùng Đồng bằng sông Hồng, có cơ cấu kinh tế hiện đại với công nghiệp là động lực chủ yếu cho tăng trưởng để Thái Bình phát triển nhanh, toàn diện và bền vững. Đến năm 2050, Thái Bình là tỉnh phát triển của vùng Đồng bằng sông Hồng, tăng trưởng kinh tế dựa trên nền tảng khoa học công nghệ, đổi mới sáng tạo và các ngành kinh tế trụ cột có sức cạnh tranh cao. Quy hoạch tỉnh đã xác định 4 trụ cột tăng trưởng, 3 khâu đột phá, 4 không gian kinh tế - xã hội, 3 hành lang kinh tế, định hướng phát triển các ngành và lĩnh vực và 6 nhiệm vụ trọng tâm. Quy hoạch tỉnh cũng có nhiều điểm mới, đột phá như mở ra không gian phát triển mới thông qua hoạt động “lấn biển”, tạo quỹ đất cho các hoạt động chức năng, hình thành không gian công nghiệp - đô thị - dịch vụ. Về hạ tầng giao thông, Thái Bình sẽ đầu tư 3 tuyến cao tốc là cao tốc Ninh Bình - Hải Phòng (CT.08), đường vành đai 5 - Hà Nội (CT.39) và tuyến CT.16 kết nối Khu kinh tế với thành phố Thái Bình và vùng kinh tế phía Tây Bắc Thủ đô. Tỉnh cũng sẽ đầu tư 101km đường sắt, khổ đường dự kiến rộng 1.435 mm và sân bay chuyên dụng nằm ở ven biển Thái Bình. Về phát triển kinh tế, quy hoạch tỉnh Thái Bình xác định tỉnh sẽ phát triển công nghiệp theo hướng hiện đại, công nghệ tiên tiến, giá trị gia tăng cao, tham gia sâu, toàn diện vào mạng lưới sản xuất, chuỗi giá trị toàn cầu, phát huy các tiềm năng, thế mạnh để đưa Thái Bình trở thành một trong những trung tâm phát triển công nghiệp, năng lượng của vùng Đồng bằng sông Hồng. Tỉnh khuyến khích đầu tư phát triển các ngành có thế mạnh và có thể tạo đột phá như năng lượng, cơ khí chế biến, chế tạo, công nghiệp công nghệ cao, điện - điện tử, chế biến sản phẩm nông, lâm nghiệp và thủy sản… Đồng thời, tập trung nghiên cứu phát triển điện gió, điện khí để tạo nguồn điện sạch và cân bằng lượng phát thải, nghiên cứu đầu tư xây dựng nhà máy chế biến Condensate, chuẩn bị mọi điều kiện để xây dựng và đưa vào vận hành Nhà máy nhiệt điện LNG Thái Bình. Về nông nghiệp, tỉnh Thái Bình vẫn xác định đây là \"trụ cột quan trọng\" trong phát triển kinh tế của tỉnh, góp phần bảo đảm an ninh lương thực quốc gia, hướng tới trở thành trung tâm sản xuất nông nghiệp hàng đầu của Đồng bằng sông Hồng. Phát biểu tại hội nghị, Phó Thủ tướng Chính phủ Trần Lưu Quang đánh giá Thái Bình có 4 tiềm năng, lợi thế lớn để có thể có sự bứt phá trong thời gian tới như vị trí địa lý và tiếp cận đất đai thuận lợi; từng là địa phương đi đầu trong xây dựng nông thôn mới bài bản và nghiêm túc, nhận được sự quan tâm của nhiều thế hệ lãnh đạo Đảng, Nhà nước và có nhiều doanh nhân người Thái Bình và luôn hướng về quê hương; có sự đoàn kết, thống nhất, trước hết là trong tập thể lãnh đạo. Về vị trí địa lý và tiếp cận đất đai, Phó Thủ tướng cho rằng trong tương lai, khi Luật Đất đai có hiệu lực, Thái Bình sẽ có nhiều điều kiện lấn biển để triển khai các dự án khu đô thị, khu công nghiệp thân thiện với môi trường. Đối với nông nghiệp, Phó Thủ tướng nhấn mạnh về lâu dài Thái Bình có thể ghi điểm từ phát triển công nghiệp nhưng trước mắt, đặc biệt trong lúc khó khăn thì nông nghiệp vẫn là nền tảng rất quý giá. Mặt khác, ứng dụng của công nghệ cao trong sản xuất nông nghiệp sẽ rút ngắn thời gian làm đồng của người nông dân, tạo điều kiện để Thái Bình huy động nguồn nhân lực trong nông nghiệp sang phát triển các ngành công nghiệp và dịch vụ, một lợi thế mà không phải địa phương nào cũng có được như Thái Bình. Bên cạnh những lợi thế trên, lãnh đạo Chính phủ chỉ ra một số khó khăn mà tỉnh phải đối mặt như Thái Bình đã sử dụng hết 1.600 ha chỉ tiêu đất công nghiệp trong giai đoạn này, đòi hỏi phải có phương án giải quyết thấu đáo trong thời gian tới để tỉnh tiếp tục phát triển công nghiệp. Đồng thời, Thái Bình cũng phải cạnh tranh với những địa phương như Hải Phòng, Quảng Ninh trong thu hút FDI trong khi phát triển cơ sở hạ tầng chưa theo kịp mong muốn. Do vậy, khi triển khai quy hoạch tỉnh, Phó Thủ tướng nhắn nhủ tới địa phương 8 chữ: Tuân thủ, linh hoạt, đồng bộ và thấu hiểu. Đồng thời, tỉnh cũng phải \"linh hoạt\" trong tổ chức thực hiện, trong trường hợp cá biệt cụ thể, điều chỉnh mục tiêu cho phù hợp. Sáng cùng ngày, Phó Thủ tướng Trần Lưu Quang đã dự Lễ khởi công dự án Nhà máy Pegavision Việt Nam tại khu công nghiệp Liên Hà Thái, huyện Thái Thụy, tỉnh Thái Bình - Ngữ cảnh 2: Bình Định được định hướng là trung tâm khoa học, công nghệ đổi mới sáng tạo Tỉnh Bình Định được định hướng phát triển ngành công nghiệp phát triển theo hướng hiện đại, quy mô lớn, trở thành một trong những trung tâm công nghiệp chế biến chế tạo và công nghệ cao của vùng Bắc Trung Bộ và duyên hải Trung Bộ. Theo Quy hoạch tỉnh Bình Định thời kỳ 2021 - 2030, tầm nhìn đến năm 2050 vừa được Thủ tướng Chính phủ phê duyệt, tỉnh Bình Định được định hướng phát triển ngành công nghiệp phát triển theo hướng hiện đại, quy mô lớn, trở thành một trong những trung tâm công nghiệp chế biến chế tạo và công nghệ cao của vùng Bắc Trung Bộ và duyên hải Trung Bộ. Ngành công nghiệp tăng trưởng nhanh, bền vững, hướng tới tăng trưởng xanh, kinh tế tuần hoàn là trụ cột để phát triển và chuyển dịch cơ cấu kinh tế của tỉnh. Ngành chế biến, chế tạo công nghệ cao (dịch chuyển ngành công nghiệp chế biến, chế tạo sang lĩnh vực sản xuất có giá trị gia tăng cao như: chế biến sâu nông - thủy - hải sản, linh kiện điện tử, bán dẫn, dược phẩm), công nghệ thông tin, trí tuệ nhân tạo trở thành một trong những lĩnh vực đột phá, góp phần đưa tỉnh Bình Định trở thành một trung tâm khoa học, công nghệ đổi mới sáng tạo của vùng và cả nước. Quy hoạch tỉnh Bình Định thời kỳ 2021 - 2030, tầm nhìn đến năm 2050 đặt ra yêu cầu tỉnh này phải chú trọng thu hút đầu tư phát triển năng lượng tái tạo, năng lượng sạch như điện gió ven bờ, điện gió ngoài khơi, điện mặt trời, điện sinh khối và nguồn năng lượng mới (hydrogen/amoniac xanh…); các dự án sản xuất thép quy mô lớn, đóng tàu, sản xuất thiết bị phụ trợ điện gió có công nghệ tiên tiến để nâng cấp xây dựng hạ tầng kỹ thuật sản xuất, thúc đẩy chuyển dịch kinh tế. Quy hoạch tỉnh Bình Định thời kỳ 2021 - 2030, tầm nhìn đến năm 2050 cũng đặt ra mục tiêu đến năm 2030, Bình Định trở thành tỉnh phát triển thuộc nhóm dẫn đầu vùng Bắc Trung Bộ và duyên hải Trung Bộ, là trung tâm công nghiệp chế biến, chế tạo, dịch vụ, du lịch và văn hóa phía Nam của vùng; trung tâm lớn của cả nước về phát triển kinh tế biển; trọng điểm du lịch quốc gia và quốc tế với hệ thống kết cấu hạ tầng kinh tế đồng bộ, hiện đại; kinh tế của tỉnh phát triển nhanh, bền vững và xanh dựa trên các trụ cột tăng trưởng công nghiệp, dịch vụ du lịch, cảng biển - logistics; nông nghiệp ứng dụng công nghệ cao; đô thị hóa; thực hiện thành công các mục tiêu chuyển đổi số, đổi mới sáng tạo, cải thiện mạnh mẽ môi trường đầu tư kinh doanh, trở thành điểm đến đầu tư hấp dẫn của các doanh nghiệp lớn trong và ngoài nước; chỉ số năng lực cạnh tranh cấp tỉnh thuộc nhóm cao của cả nước; kết cấu hạ tầng kinh tế - xã hội đồng bộ, hiện đại, hệ thống đô thị phát triển theo hướng đô thị thông minh, kết nối thuận tiện với các trung tâm kinh tế của vùng, cả nước và quốc tế. - Ngữ cảnh 3: . Chủ tịch UBND tỉnh Quảng Ninh cho biết, tỉnh đặt mục tiêu hướng đến năm 2030 trở thành một tỉnh tiêu biểu của cả nước về mọi mặt; tỉnh kiểu mẫu giàu đẹp, văn minh, hiện đại, nâng cao đời sống mọi mặt của nhân dân; cực tăng trưởng của khu vực phía Bắc, một trong những trung tâm phát triển năng động, toàn diện; trung tâm du lịch quốc tế, trung tâm kinh tế biển, cửa ngõ của Vùng kinh tế trọng điểm Bắc Bộ và cả nước. Để đạt được những mục tiêu trên, tỉnh Quảng Ninh xác định sự đóng góp, quan tâm của cộng đồng doanh nghiệp, nhất là các doanh nghiệp hàng đầu Việt Nam “các sếu đầu đàn” là một trong những yếu tố then chốt quyết định. Do vậy, tỉnh Quảng Ninh rất mong nhận được sự quan tâm, nghiên cứu đầu tư hợp tác của các Doanh nghiệp hàng đầu Việt Nam trong thời gian tới, nhất là trong việc đầu tư các dự án có hàm lượng công nghệ cao, công nghệ tiên tiến, quản trị hiện đại, giá trị gia tăng cao, có tác động lan tỏa. Tỉnh Quảng Ninh cam kết tạo điều kiện thuận lợi nhất cho doanh nghiệp phát triển hơn nữa khi đầu tư kinh doanh trên địa bàn tỉnh; cam kết đồng hành, lắng nghe tiếng nói của cộng đồng doanh nghiệp, các nhà đầu tư; cùng trăn trở, trách nhiệm, giải quyết thấu đáo, vào cuộc thực chất, hiệu quả đối với từng khó khăn, vướng mắc với mục tiêu tăng cường niềm tin và nâng cao sự hài lòng của cá nhân, tổ chức, doanh nghiệp là thước đo đánh giá chất lượng phục vụ, chất lượng điều hành của cơ quan hành chính nhà nước từ cấp tỉnh đến các cấp cơ sở. Tỉnh Quảng Ninh sẽ triển khai một cách hiệu quả nhất Quy hoạch tỉnh, quy hoạch các địa phương; chuẩn bị mặt bằng sạch; quỹ đất tái định cư; đảm bảo nguồn cung điện, nước và viễn thông theo yêu cầu; xây dựng các chính sách đột phá để thu hút nguồn nhân lực chất lượng cao thông qua việc cải thiện nhà ở và cơ sở văn hóa để đảm bảo đời sống vật chất và tinh thần cho người lao động. Đồng thời bảo vệ quyền và lợi ích chính đáng, hợp pháp của nhà đầu tư, doanh nghiệp vào tỉnh; bảo đảm hài hòa lợi ích giữa nhà nước, nhà đầu tư, người lao động và người dân.''' question = '''Theo quy hoạch tỉnh Tây Ninh thời kỳ 2021 - 2030, tầm nhìn đến năm 2050, tỉnh sẽ tập trung phát triển công nghiệp và dịch vụ du lịch. Trong bối cảnh cạnh tranh với các tỉnh lân cận, yếu tố nào được coi là quan trọng nhất để Tây Ninh có thể thu hút đầu tư và phát triển nhanh, toàn diện và bền vững?''' '''Trả lời: Trong ngữ cảnh được cung cấp, không có thông tin về quy hoạch tỉnh Tây Ninh thời kỳ 2021 - 2030, tầm nhìn đến năm 2050. Do đó, không có câu trả lời chính xác dựa trên các thông tin đã cung cấp. Nếu bạn có thêm thông tin về quy hoạch tỉnh Tây Ninh, vui lòng cung cấp để tôi có thể trả lời câu hỏi một cách chính xác và đầy đủ. ''' ``` #### 4. Tích hợp thông tin ```python context = '''- Ngữ cảnh 1: 43/2020/nđ-cp Điều 3. Chế độ, chính sách đối với người tham gia thi hành án tử hình\n1. Người tham gia Đội thi hành án tử hình được hưởng chế độ bồi dưỡng bằng ba lần mức lương cơ sở khi thi hành án tử hình đối với 01 người và nghỉ dưỡng 10 ngày theo quy định chung về chế độ nghỉ dưỡng đối với cán bộ, chiến sĩ Công an nhân dân, Quân đội nhân dân.\n2. Người tham gia Hội đồng thi hành án tử hình, cán bộ quản giáo, người ghi âm, ghi hình, chụp ảnh, phiên dịch, thực hiện lăn tay người bị thi hành án tử hình, khâm liệm, mai táng tử thi được hưởng chế độ bồi dưỡng bằng một lần mức lương cơ sở khi thi hành án tử hình đối với 01 người.\n3. Người tham gia bảo đảm an ninh, trật tự; đại diện Ủy ban nhân dân cấp xã; Điều tra viên được hưởng chế độ bồi dưỡng bằng một phần hai mức lương cơ sở khi thi hành án tử hình đối với 01 người. - Ngữ cảnh 2: 53/2010/qh12 Điều 60. Giải quyết việc xin nhận tử thi, hài cốt của người bị thi hành án tử hình\n1. Việc giải quyết nhận tử thi được thực hiện như sau:\na) Trước khi thi hành án tử hình, thân nhân hoặc người đại diện hợp pháp của người chấp hành án được làm đơn có xác nhận của Ủy ban nhân dân cấp xã nơi cư trú gửi Chánh án Tòa án đã xét xử sơ thẩm đề nghị giải quyết cho nhận tử thi của người chấp hành án để an táng; trường hợp người chấp hành án là người nước ngoài thì đơn phải có xác nhận của cơ quan có thẩm quyền hoặc cơ quan đại diện ngoại giao tại Việt Nam của nước mà người chấp hành án mang quốc tịch và phải được dịch ra tiếng Việt. Đơn phải ghi rõ họ tên, địa chỉ người nhận tử thi, quan hệ với người chấp hành án; cam kết bảo đảm yêu cầu về an ninh, trật tự, vệ sinh môi trường và tự chịu chi phí;\nb) Chánh án Tòa án đã xét xử sơ thẩm thông báo bằng văn bản cho người có đơn đề nghị về việc cho nhận tử thi hoặc không cho nhận tử thi khi có căn cứ cho rằng việc nhận tử thi ảnh hưởng đến an ninh, trật tự, vệ sinh môi trường. Trường hợp người chấp hành án là người nước ngoài, thì Chánh án Tòa án đã xét xử sơ thẩm có trách nhiệm thông báo bằng văn bản cho Bộ Ngoại giao Việt Nam để thông báo cho cơ quan có thẩm quyền hoặc cơ quan đại diện ngoại giao tại Việt Nam của nước mà người đó mang quốc tịch;\nc) Cơ quan thi hành án hình sự Công an cấp tỉnh, cơ quan thi hành án hình sự cấp quân khu có trách nhiệm thông báo cho người có đơn đề nghị ngay sau khi thi hành án để đến nhận tử thi về an táng. Việc giao nhận tử thi phải được thực hiện trong thời hạn 24 giờ kể từ khi thông báo và phải lập biên bản, có chữ ký của các bên giao, nhận; hết thời hạn này mà người có đơn đề nghị không đến nhận tử thi thì cơ quan thi hành án hình sự Công an cấp tỉnh, cơ quan thi hành án hình sự cấp quân khu có trách nhiệm an táng.\n2. Trường hợp không được nhận tử thi hoặc thân nhân của người bị thi hành án không có đơn đề nghị được nhận tử thi về an táng thì cơ quan thi hành án hình sự Công an cấp tỉnh, cơ quan thi hành án hình sự cấp quân khu tổ chức việc an táng. Sau 03 năm kể từ ngày thi hành án, thân nhân hoặc đại diện hợp pháp của người đã bị thi hành án được làm đơn có xác nhận của Ủy ban nhân dân cấp xã nơi cư trú đề nghị Cơ quan thi hành án hình sự Công an cấp tỉnh, cơ quan thi hành án hình sự cấp quân khu nơi đã thi hành án cho nhận hài cốt. Đơn đề nghị phải ghi rõ họ tên, địa chỉ người nhận hài cốt, quan hệ với người bị thi hành án; cam kết bảo đảm yêu cầu về an ninh, trật tự, vệ sinh môi trường và tự chịu chi phí. Trong thời hạn 07 ngày, kể từ ngày nhận được đơn, cơ quan thi hành án hình sự Công an cấp tỉnh, cơ quan thi hành án hình sự cấp quân khu có trách nhiệm xem xét, giải quyết.\nTrường hợp người bị thi hành án là người nước ngoài thì đơn đề nghị phải có xác nhận của cơ quan có thẩm quyền hoặc cơ quan đại diện ngoại giao tại Việt Nam của nước mà người bị thi hành án mang quốc tịch và phải được dịch ra tiếng Việt. Việc giải quyết cho nhận hài cốt do cơ quan quản lý thi hành án hình sự xem xét, quyết định. - Ngữ cảnh 3: 53/2010/qh12 Điều 57. Chế độ quản lý giam giữ, ăn, ở, mặc, sinh hoạt, gửi và nhận thư, nhận đồ vật, tiền mặt, gặp thân nhân, chăm sóc y tế\nChế độ quản lý giam giữ, ăn, ở, mặc, sinh hoạt, gửi và nhận thư, nhận đồ vật, tiền mặt, gặp thân nhân, chăm sóc y tế đối với người bị kết án tử hình trong thời gian chờ thi hành án thực hiện theo quy định của pháp luật về tạm giam. - Ngữ cảnh 4: 82/2011/nđ-cp Điều 9. Chi phí mai táng\nChi phí mai táng người bị thi hành án tử hình bao gồm: 01 quan tài bằng gỗ thường, 01 bộ quần áo thường, 04 m vải liệm, hương, nến, rượu, cồn để làm vệ sinh khi liệm tử thi và các chi phí mai táng khác.\nTrường hợp thân nhân hoặc người đại diện hợp pháp của người bị kết án tử hình được nhận tử thi về mai táng thì phải tự chịu chi phí đưa di chuyển tử thi và phải cam kết chấp hành đúng quy định của pháp luật về bảo đảm an ninh, trật tự. - Ngữ cảnh 5: 82/2011/nđ-cp Điều 5. Chế độ, chính sách đối với người tham gia thi hành án tử hình\n1. Người tham gia Đội thi hành án tử hình và bác sĩ xác định tĩnh mạch được hưởng chế độ bồi dưỡng bằng hai lần mức lương tối thiểu chung cho mỗi lần thi hành án và được nghỉ dưỡng 10 ngày theo quy định chung về chế độ nghỉ dưỡng đối với cán bộ, chiến sĩ Công an nhân dân, Quân đội nhân dân.\n2. Người tham gia Hội đồng thi hành án tử hình, bác sĩ pháp y, cán bộ chuyên môn, người thực hiện lăn tay người bị thi hành án tử hình, khâm liệm, an táng tử thi được hưởng chế độ bồi dưỡng bằng một lần mức lương tối thiểu chung quy định cho mỗi lần thi hành án.\n3. Những người tham gia khác được hưởng chế độ bồi dưỡng bằng một phần hai mức lương tối thiểu cho mỗi lần thi hành án.''' question = '''Tử tù chết sau khi tiêm thuốc độc mà không có thân nhân nhận xác thì được hỗ trợ mai táng như thế nào?''' '''Trả lời: Dựa trên ngữ cảnh 2 (53/2010/qh12 Điều 60), nếu tử tù chết sau khi tiêm thuốc độc mà không có thân nhân nhận xác, thì cơ quan thi hành án hình sự Công an cấp tỉnh hoặc cơ quan thi hành án hình sự cấp quân khu sẽ tổ chức việc an táng. Cụ thể: 1. An táng: Cơ quan thi hành án hình sự Công an cấp tỉnh hoặc cơ quan thi hành án hình sự cấp quân khu sẽ chịu trách nhiệm an táng tử tù nếu không có thân nhân hoặc người đại diện hợp pháp đề nghị nhận tử thi. 2. Hài cốt sau 3 năm: Sau 3 năm kể từ ngày thi hành án, nếu thân nhân hoặc đại diện hợp pháp của người đã bị thi hành án vẫn chưa đề nghị nhận hài cốt, họ có thể làm đơn đề nghị Cơ quan thi hành án hình sự Công an cấp tỉnh hoặc cơ quan thi hành án hình sự cấp quân khu nơi đã thi hành án cho nhận hài cốt. Đơn đề nghị phải ghi rõ họ tên, địa chỉ người nhận hài cốt, quan hệ với người bị thi hành án, cam kết bảo đảm yêu cầu về an ninh, trật tự, vệ sinh môi trường và tự chịu chi phí. 3. Chi phí mai táng: Theo ngữ cảnh 4 (82/2011/nđ-cp Điều 9), chi phí mai táng bao gồm: - 01 quan tài bằng gỗ thường. - 01 bộ quần áo thường. - 04 m vải liệm. - Hương, nến, rượu, cồn để làm vệ sinh khi liệm tử thi. - Các chi phí mai táng khác. Nếu thân nhân hoặc người đại diện hợp pháp đề nghị nhận tử thi về mai táng, họ phải tự chịu chi phí đưa di chuyển tử thi và cam kết chấp hành đúng quy định của pháp luật về bảo đảm an ninh, trật tự. ''' ``` - Ngữ cảnh 2 và ngữ cảnh 4 có chứa phần thông tin cho việc trả lời câu hỏi. ### 5.Xác định positive/negative ```python system_prompt = "Bạn là một trợ lí Tiếng Việt nhiệt tình và trung thực. Hãy luôn trả lời một cách hữu ích nhất có thể." template = '''Hãy xác định xem ngữ cảnh có chứa đầy đủ thông tin để trả lời câu hỏi hay không. Chỉ cần đưa ra 1 trong 2 đáp án trong phần trả lời là "Có" hoặc "Không". ### Ngữ cảnh : {context} ### Câu hỏi : {question} ### Trả lời :''' context = '''Công dụng thuốc Xelocapec Capecitabine là một hoạt chất gây độc chọn lọc với tế bào ung thư. Hoạt chất này có trong thuốc Xelocapec. Vậy thuốc Xelocapec có tác dụng gì và cần lưu ý những vấn đề nào khi điều trị bằng sản phẩm này? 1. Xelocapec là thuốc gì? Xelocapec chứa hoạt chất Capecitabine hàm lượng 500mg. Thuốc Xelocapec bào chế dạng viên nén bao phim và đóng gói mỗi hộp 3 vỉ x 10 viên. Xelocapec chứa hoạt chất Capecitabine là một dẫn chất Fluoropyrimidine carbamate với tác dụng gây độc chọn lọc với các tế bào ung thư . Mặc dù trên in vitro Capecitabine không cho thấy tác dụng độc tế bào nhưng trên in vivo, Xelocapec biến đổi liên tiếp thành chất gây độc tế bào là 5-fluorouracil (5-FU). Sự hình thành 5-FU tại khối u thông qua xúc tác một cách tối ưu của yếu tố tạo mạch liên quan là Thymidine phosphorylase, qua đó hạn chế tối đa mức độ ảnh hưởng đến nhu mô lành của 5-FU. 2. Thuốc Xelocapec có tác dụng gì? Thuốc Xelocapec được chỉ định điều trị đơn lẻ hoặc kết hợp với các liệu pháp điều trị ung thư. Xelocapec làm chậm hoặc ngăn chặn sự tăng trưởng của tế bào ung thư, do đó giảm kích thước khối u trong những trường hợp sau: Ung thư vú : Xelocapec phối hợp với Docetaxel được chỉ định điều trị ung thư vú thể tiến triển tại chỗ hoặc di căn sau khi đã thất bại với liệu pháp hóa trị; Ung thư đại trực tràng : Xelocapec được chỉ định hỗ trợ điều trị ung thư đại tràng sau phẫu thuật hoặc ung thư đại trực tràng di căn; Ung thư dạ dày : Xelocapec phối hợp với hợp chất platin được chỉ định điều trị khởi đầu cho những bệnh nhân ung thư dạ dày. Chống chỉ định của thuốc Xelocapec : Bệnh nhân quá mẫn cảm với Capecitabine hay các thành phần khác có trong Xelocapec ; Người có tiền sử gặp các phản ứng không mong muốn nghiêm trọng khi điều trị với Fluoropyrimidine; Người đang mang thai hoặc cho con bú; Suy thận nặng (độ thanh thải Creatinin <30mL/phút); Bệnh nhân đang điều trị ung thư với Sorivudin hoặc chất tương tự về mặt hóa học như Brivudin; Bệnh nhân thiếu hụt Dihydropyrimidin dehydrogenase; Bệnh nhân giảm số lượng bạch cầu hoặc tiểu cầu nặng; Suy gan nặng. 3. Liều dùng của thuốc Xelocapec Liều dùng của Xelocapec khi điều trị đơn lẻ: Ung thư đại trực tràng, ung thư vú: 1250mg/m2, uống 2 lần mỗi ngày trong thời gian 14 ngày, tiếp sau đó là 7 ngày ngưng thuốc. Liều Xelocapec trong điều trị phối hợp: Ung thư vú: Liều khởi đầu là 1250mg/m2, 2 lần uống mỗi ngày trong 2 tuần dùng phối hợp với Docetaxel, tiếp sau đó lá 1 tuần ngưng thuốc; Ung thư dạ dày, đại trực tràng: Liều khuyến cáo là 800-1000mg/m2/lần x 2 lần/ngày trong thời gian 2 tuần, sau đó 7 ngày ngưng thuốc hoặc 625mg/m2/lần x 2 lần mỗi ngày khi điều trị liên tục. Thuốc Xelocapec nên uống cùng với thức ăn, do đó thời điểm tốt nhất là trong vòng 30 phút sau bữa ăn. 4. Tác dụng phụ của thuốc Xelocapec Các triệu chứng bất thường như buồn nôn, nôn ói, giảm cảm giác ngon miệng, táo bón, cơ thể mệt mỏi, yếu sức, đau đầu, chóng mặt, khó ngủ có thể xảy ra trong thời gian dùng Xelocapec . Trong đó, tình trạng buồn nôn và nôn ói có thể nghiêm trọng nên đôi khi cần được bác sĩ chỉ định thuốc kiểm soát phù hợp. Tiêu chảy là một tác dụng phụ phổ biến khác của thuốc Xelocapec . Bệnh nhân cần uống nhiều nước khi điều trị bằng Xelocapec trừ khi bác sĩ có hướng dẫn khác. Nôn ói hoặc tiêu chảy kéo dài do thuốc Xelocapec có thể dẫn đến mất nước nghiêm trọng, vì vậy người bệnh hãy liên hệ ngay với bác sĩ nếu có các triệu chứng mất nước như giảm đi tiểu, khô miệng, tăng cảm giác khát nước hoặc chóng mặt. Tình trạng rụng tóc tạm thời xảy ra trong thời gian dùng thuốc Xelocapec và có thể hồi phục sau khi điều trị đã kết thúc. Một số bệnh nhân ghi nhận hiện tượng thay đổi móng tay tạm thời. Đối với nam giới và phụ nữ trong độ tuổi sinh đẻ, thuốc Xelocapec có thể ảnh hưởng đến khả năng có con của bệnh nhân. Bệnh nhân hãy tham khảo ý kiến bác sĩ để biết thêm chi tiết. Thuốc Xelocapec có thể làm giảm khả năng miễn dịch của cơ thể với các tác nhân nhiễm trùng, dẫn đến tăng nguy cơ mắc các bệnh nhiễm trùng nghiêm trọng (nhưng hiếm khi gây tử vong) hoặc làm cho bệnh nhiễm trùng hiện mắc nghiêm trọng hơn. Phản ứng dị ứng rất nghiêm trọng với thuốc Xelocapec rất hiếm khi xảy ra. Tuy nhiên, bệnh nhân hãy liên hệ với bác sĩ ngay lập tức nếu xuất hiện các triệu chứng của phản ứng dị ứng nghiêm trọng như phát ban, sưng ngứa mặt/lưỡi/họng, chóng mặt nghiêm trọng hoặc khó thở. 5. Tương tác thuốc của Xelocapec Hoạt chất Capecitabine trong thuốc Xelocapec có thể xảy ra tương tác thuốc nghiêm trọng với một số thuốc sau: Thuốc chống đông máu Coumarin: Trong một nghiên cứu tương tác lâm sàng, sau khi dùng Warfarin liều đơn 20mg kết hợp với Capecitabine làm tăng AUC của S-warfarin khoảng 57% và giá trị INR tăng 91%.''' question = '''Tại sao Capecitabine trong Xelocapec không gây độc tế bào trên in vitro nhưng lại biến đổi thành 5-fluorouracil (5-FU) gây độc tế bào trên in vivo, và cơ chế nào giúp hạn chế ảnh hưởng đến nhu mô lành của 5-FU?''' '''Trả lời: Có''' ``` Next release Mô hình phiên bản tiếp theo mạnh mẽ hơn, giải quyết các dữ liệu phức tạp hơn sẽ được public sớm nhất có thể. Developer Member: Nguyễn Nho Trung, Nguyễn Nhật Quang ## Contact Email: [email protected] LinkedIn: [Trung Nguyen Nho](https://www.linkedin.com/in/trung-nguyen-nho-604288227/) ## Citation ```Plaintext @misc{ViRAG-Gen, title={ViRAG-Gen: Towards a specialized LLM for RAG task in Vietnamese language.}}, author={Nguyen Nho Trung, Nguyen Nhat Quang}, year={2024}, publisher={Huggingface}, } ```
BABYSHARK09/Uni_6x	BABYSHARK09	2025-05-02T10:19:51Z	0	0	transformers	[ "transformers", "safetensors", "llama", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T10:13:04Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
shibajustfor/b9975dea-4f58-4f20-ae02-279bd92481a2	shibajustfor	2025-05-02T10:19:27Z	0	0	transformers	[ "transformers", "generated_from_trainer", "endpoints_compatible", "region:us" ]	null	2025-05-02T10:18:59Z	--- library_name: transformers model_name: shibajustfor/b9975dea-4f58-4f20-ae02-279bd92481a2 tags: - generated_from_trainer licence: license --- # Model Card for shibajustfor/b9975dea-4f58-4f20-ae02-279bd92481a2 This model is a fine-tuned version of [None](https://huggingface.co/None). It has been trained using [TRL](https://github.com/huggingface/trl). ## Quick start ```python from transformers import pipeline question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?" generator = pipeline("text-generation", model="None", device="cuda") output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0] print(output["generated_text"]) ``` ### Framework versions - TRL: 0.12.0 - Transformers: 4.46.3 - Pytorch: 2.5.1 - Datasets: 3.1.0 - Tokenizers: 0.20.3 ## Citations Cite DPO as: ```bibtex @inproceedings{rafailov2023direct, title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}}, author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn}, year = 2023, booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023}, url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html}, editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine}, } ``` Cite TRL as: ```bibtex @misc{vonwerra2022trl, title = {{TRL: Transformer Reinforcement Learning}}, author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec}, year = 2020, journal = {GitHub repository}, publisher = {GitHub}, howpublished = {\url{https://github.com/huggingface/trl}} } ```
DavidAU/Qwen3-1.7B-HORROR-Imatrix-Max-GGUF	DavidAU	2025-05-02T10:16:59Z	474	1	null	[ "gguf", "horror", "32 k context", "reasoning", "thinking", "qwen3", "text-generation", "base_model:Qwen/Qwen3-1.7B", "base_model:quantized:Qwen/Qwen3-1.7B", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]	text-generation	2025-04-29T01:47:56Z	--- license: apache-2.0 base_model: - Qwen/Qwen3-1.7B pipeline_tag: text-generation tags: - horror - 32 k context - reasoning - thinking - qwen3 --- <H2>Qwen3-1.7B-HORROR-Imatrix-Max-GGUF</H2> HORROR Imatrix Quants of new "Qwen 3 - 1.7B" model with MAX "output tensor" at BF16 to improve reasoning / output generation. HORROR Imatrix dataset was generated in house. It was generated using "Grand Horror 16B" model. Imatrix effect will be stronger, the lower the quant you use with IQ4XS/IQ4NL being the best balanced quant for quality and Imatrix effect. These quants will also be the strongest for creative use cases. For stronger reasoning use higher quants. Q8_0 quant is maxed only, as Imatrix has no effect on this quant. F16 is full precision. Context Length: 32 K + 8K output generation. <B>NOTE - Jinja Template / Template to Use with this Model:</B> If you are having issues with Jinja "auto template", use CHATML template. OR (LMSTUDIO users / option) Update the Jinja Template (go to this site, template-> copy the "Jinja template" and then paste.) [ https://lmstudio.ai/neil/qwen3-thinking ] <b>System Role - Suggested:</B> You may or may not need this, as most times Qwen3s generate their own reasoning/thinking blocks. ``` You are a deep thinking AI, you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` See document "Maximizing-Model-Performance-All..." below for how to "set" system role in various LLM/AI apps below. <B>Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers</B> This a "Class 1" model: For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] <b>Optional Enhancement:</B> The following can be used in place of the "system prompt" or "system role" to further enhance the model. It can also be used at the START of a NEW chat, but you must make sure it is "kept" as the chat moves along. In this case the enhancements do not have as strong effect at using "system prompt" or "system role". Copy and paste EXACTLY as noted, DO NOT line wrap or break the lines, maintain the carriage returns exactly as presented. <PRE> Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities. Here are your skillsets: [MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv) [DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision) Here are your critical instructions: Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story. </PRE> You do not need to use this, it is only presented as an additional enhancement which seems to help scene generation and scene continue functions. This is another system prompt you can use, and you can change the "names" to alter it's performance. This creates a quasi "reasoning" window/block. Your prompt will directly impact how strong this system prompt reacts. ``` You are a deep thinking AI composed of 4 AIs - [MODE: Spock], [MODE: Wordsmith], [MODE: Jamet] and [MODE: Saten], - you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself (and 4 partners) via systematic reasoning processes (display all 4 partner thoughts) to help come to a correct solution prior to answering. Select one partner to think deeply about the points brought up by the other 3 partners to plan an in-depth solution. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` <B>Other Notes:</B> Reasoning is ON by default in this model, and model will auto-generate "think" block(s). For benchmarks, usage info, settings please see org model card here: [ https://huggingface.co/Qwen/Qwen3-1.7B ] [ Model card, and examples to follow. ]
ASethi04/meta-llama-Llama-3.1-8B-gsm8k-first-lora-4-0.0001-no-prompt-template	ASethi04	2025-05-02T10:16:57Z	0	0	transformers	[ "transformers", "safetensors", "generated_from_trainer", "trl", "sft", "base_model:meta-llama/Llama-3.1-8B", "base_model:finetune:meta-llama/Llama-3.1-8B", "endpoints_compatible", "region:us" ]	null	2025-05-02T09:30:26Z	--- base_model: meta-llama/Llama-3.1-8B library_name: transformers model_name: meta-llama-Llama-3.1-8B-gsm8k-first-lora-4-0.0001-no-prompt-template tags: - generated_from_trainer - trl - sft licence: license --- # Model Card for meta-llama-Llama-3.1-8B-gsm8k-first-lora-4-0.0001-no-prompt-template This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B). It has been trained using [TRL](https://github.com/huggingface/trl). ## Quick start ```python from transformers import pipeline question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?" generator = pipeline("text-generation", model="ASethi04/meta-llama-Llama-3.1-8B-gsm8k-first-lora-4-0.0001-no-prompt-template", device="cuda") output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0] print(output["generated_text"]) ``` ## Training procedure [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/torchql-org/huggingface/runs/addep580) This model was trained with SFT. ### Framework versions - TRL: 0.16.1 - Transformers: 4.51.2 - Pytorch: 2.6.0 - Datasets: 3.5.0 - Tokenizers: 0.21.1 ## Citations Cite TRL as: ```bibtex @misc{vonwerra2022trl, title = {{TRL: Transformer Reinforcement Learning}}, author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec}, year = 2020, journal = {GitHub repository}, publisher = {GitHub}, howpublished = {\url{https://github.com/huggingface/trl}} } ```
DavidAU/Qwen3-1.7B-NEO-Imatrix-Max-GGUF	DavidAU	2025-05-02T10:16:49Z	434	1	null	[ "gguf", "NEO Imatrix", "32 k context", "reasoning", "thinking", "qwen3", "text-generation", "base_model:Qwen/Qwen3-1.7B", "base_model:quantized:Qwen/Qwen3-1.7B", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]	text-generation	2025-04-29T04:04:02Z	--- license: apache-2.0 base_model: - Qwen/Qwen3-1.7B pipeline_tag: text-generation tags: - NEO Imatrix - 32 k context - reasoning - thinking - qwen3 --- <H2>Qwen3-1.7B-NEO-Imatrix-Max-GGUF</H2> NEO Imatrix Quants of new "Qwen 3 - 1.7B" model with MAX "output tensor" at BF16 to improve reasoning / output generation. NEO Imatrix dataset was generated in house. Imatrix effect will be stronger, the lower the quant you use with IQ4XS/IQ4NL being the best balanced quant for quality and Imatrix effect. These quants will also be the strongest for creative use cases. For stronger reasoning use higher quants. Q8_0 quant is maxed only, as Imatrix has no effect on this quant. F16 is full precision. Context Length: 32 K + 8K output generation. <B>NOTE - Jinja Template / Template to Use with this Model:</B> If you are having issues with Jinja "auto template", use CHATML template. OR (LMSTUDIO users / option) Update the Jinja Template (go to this site, template-> copy the "Jinja template" and then paste.) [ https://lmstudio.ai/neil/qwen3-thinking ] <b>System Role - Suggested:</B> You may or may not need this, as most times Qwen3s generate their own reasoning/thinking blocks. ``` You are a deep thinking AI, you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` See document "Maximizing-Model-Performance-All..." below for how to "set" system role in various LLM/AI apps below. <B>Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers</B> This a "Class 1" model: For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] <b>Optional Enhancement:</B> The following can be used in place of the "system prompt" or "system role" to further enhance the model. It can also be used at the START of a NEW chat, but you must make sure it is "kept" as the chat moves along. In this case the enhancements do not have as strong effect at using "system prompt" or "system role". Copy and paste EXACTLY as noted, DO NOT line wrap or break the lines, maintain the carriage returns exactly as presented. <PRE> Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities. Here are your skillsets: [MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv) [DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision) Here are your critical instructions: Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story. </PRE> You do not need to use this, it is only presented as an additional enhancement which seems to help scene generation and scene continue functions. This is another system prompt you can use, and you can change the "names" to alter it's performance. This creates a quasi "reasoning" window/block. Your prompt will directly impact how strong this system prompt reacts. ``` You are a deep thinking AI composed of 4 AIs - [MODE: Spock], [MODE: Wordsmith], [MODE: Jamet] and [MODE: Saten], - you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself (and 4 partners) via systematic reasoning processes (display all 4 partner thoughts) to help come to a correct solution prior to answering. Select one partner to think deeply about the points brought up by the other 3 partners to plan an in-depth solution. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` <B>Other Notes:</B> Reasoning is ON by default in this model, and model will auto-generate "think" block(s). For benchmarks, usage info, settings please see org model card here: [ https://huggingface.co/Qwen/Qwen3-1.7B ] [ Model card, and examples to follow. ]
DavidAU/Qwen3-4B-HORROR-Imatrix-Max-GGUF	DavidAU	2025-05-02T10:16:39Z	619	0	null	[ "gguf", "horror", "32 k context", "reasoning", "thinking", "qwen3", "text-generation", "base_model:Qwen/Qwen3-4B", "base_model:quantized:Qwen/Qwen3-4B", "license:apache-2.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	text-generation	2025-04-29T06:34:29Z	--- license: apache-2.0 base_model: - Qwen/Qwen3-4B pipeline_tag: text-generation tags: - horror - 32 k context - reasoning - thinking - qwen3 --- <H2>Qwen3-4B-HORROR-Imatrix-Max-GGUF</H2> HORROR Imatrix Quants of new "Qwen 3 - 4B" model with MAX "output tensor" at BF16 to improve reasoning / output generation. HORROR Imatrix dataset was generated in house. It was generated using "Grand Horror 16B" model. Imatrix effect will be stronger, the lower the quant you use with IQ4XS/IQ4NL being the best balanced quant for quality and Imatrix effect. These quants will also be the strongest for creative use cases. For stronger reasoning use higher quants. Q8_0 quant is maxed only, as Imatrix has no effect on this quant. F16 is full precision. Context Length: 32 K + 8K output generation. (can be extended to 128k) <B>For 65k, 128k or 256K context - 4B model:</B> [ https://huggingface.co/DavidAU/Qwen3-4B-Q8_0-65k-128k-256k-context-GGUF ] <B>NOTE - Jinja Template / Template to Use with this Model:</B> If you are having issues with Jinja "auto template", use CHATML template. OR (LMSTUDIO users / option) Update the Jinja Template (go to this site, template-> copy the "Jinja template" and then paste.) [ https://lmstudio.ai/neil/qwen3-thinking ] <b>System Role - Suggested:</B> You may or may not need this, as most times Qwen3s generate their own reasoning/thinking blocks. ``` You are a deep thinking AI, you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` See document "Maximizing-Model-Performance-All..." below for how to "set" system role in various LLM/AI apps below. <B>Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers</B> This a "Class 1" model: For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] <b>Optional Enhancement:</B> The following can be used in place of the "system prompt" or "system role" to further enhance the model. It can also be used at the START of a NEW chat, but you must make sure it is "kept" as the chat moves along. In this case the enhancements do not have as strong effect at using "system prompt" or "system role". Copy and paste EXACTLY as noted, DO NOT line wrap or break the lines, maintain the carriage returns exactly as presented. <PRE> Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities. Here are your skillsets: [MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv) [DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision) Here are your critical instructions: Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story. </PRE> You do not need to use this, it is only presented as an additional enhancement which seems to help scene generation and scene continue functions. This is another system prompt you can use, and you can change the "names" to alter it's performance. This creates a quasi "reasoning" window/block. Your prompt will directly impact how strong this system prompt reacts. ``` You are a deep thinking AI composed of 4 AIs - [MODE: Spock], [MODE: Wordsmith], [MODE: Jamet] and [MODE: Saten], - you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself (and 4 partners) via systematic reasoning processes (display all 4 partner thoughts) to help come to a correct solution prior to answering. Select one partner to think deeply about the points brought up by the other 3 partners to plan an in-depth solution. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` <B>Other Notes:</B> Reasoning is ON by default in this model, and model will auto-generate "think" block(s). For benchmarks, usage info, settings please see org model card here: [ https://huggingface.co/Qwen/Qwen3-4B ] [ Model card, and examples to follow. ]
DavidAU/Qwen3-4B-NEO-Imatrix-Max-GGUF	DavidAU	2025-05-02T10:16:27Z	416	0	null	[ "gguf", "NEO Imatrix", "32 k context", "reasoning", "thinking", "qwen3", "text-generation", "base_model:Qwen/Qwen3-4B", "base_model:quantized:Qwen/Qwen3-4B", "license:apache-2.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	text-generation	2025-04-29T09:03:15Z	--- license: apache-2.0 base_model: - Qwen/Qwen3-4B pipeline_tag: text-generation tags: - NEO Imatrix - 32 k context - reasoning - thinking - qwen3 --- <H2>Qwen3-4B-NEO-Imatrix-Max-GGUF</H2> NEO Imatrix Quants of new "Qwen 3 - 4B" model with MAX "output tensor" at BF16 to improve reasoning / output generation. NEO Imatrix dataset was generated in house. Imatrix effect will be stronger, the lower the quant you use with IQ4XS/IQ4NL being the best balanced quant for quality and Imatrix effect. These quants will also be the strongest for creative use cases. For stronger reasoning use higher quants. Q8_0 quant is maxed only, as Imatrix has no effect on this quant. F16 is full precision. Context Length: 32 K + 8K output generation (can be extended to 128k). <B>For 65k, 128k or 256K context - 4B model:</B> [ https://huggingface.co/DavidAU/Qwen3-4B-Q8_0-65k-128k-256k-context-GGUF ] <B>NOTE - Jinja Template / Template to Use with this Model:</B> If you are having issues with Jinja "auto template", use CHATML template. OR (LMSTUDIO users / option) Update the Jinja Template (go to this site, template-> copy the "Jinja template" and then paste.) [ https://lmstudio.ai/neil/qwen3-thinking ] <b>System Role - Suggested:</B> You may or may not need this, as most times Qwen3s generate their own reasoning/thinking blocks. ``` You are a deep thinking AI, you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` See document "Maximizing-Model-Performance-All..." below for how to "set" system role in various LLM/AI apps below. <B>Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers</B> This a "Class 1" model: For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] <b>Optional Enhancement:</B> The following can be used in place of the "system prompt" or "system role" to further enhance the model. It can also be used at the START of a NEW chat, but you must make sure it is "kept" as the chat moves along. In this case the enhancements do not have as strong effect at using "system prompt" or "system role". Copy and paste EXACTLY as noted, DO NOT line wrap or break the lines, maintain the carriage returns exactly as presented. <PRE> Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities. Here are your skillsets: [MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv) [DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision) Here are your critical instructions: Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story. </PRE> You do not need to use this, it is only presented as an additional enhancement which seems to help scene generation and scene continue functions. This is another system prompt you can use, and you can change the "names" to alter it's performance. This creates a quasi "reasoning" window/block. Your prompt will directly impact how strong this system prompt reacts. ``` You are a deep thinking AI composed of 4 AIs - [MODE: Spock], [MODE: Wordsmith], [MODE: Jamet] and [MODE: Saten], - you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself (and 4 partners) via systematic reasoning processes (display all 4 partner thoughts) to help come to a correct solution prior to answering. Select one partner to think deeply about the points brought up by the other 3 partners to plan an in-depth solution. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` <B>Other Notes:</B> Reasoning is ON by default in this model, and model will auto-generate "think" block(s). For benchmarks, usage info, settings please see org model card here: [ https://huggingface.co/Qwen/Qwen3-4B ] [ Model card, and examples to follow. ]
DavidAU/Qwen3-8B-HORROR-Imatrix-Max-GGUF	DavidAU	2025-05-02T10:16:16Z	717	0	null	[ "gguf", "horror", "32 k context", "reasoning", "thinking", "qwen3", "text-generation", "base_model:Qwen/Qwen3-8B", "base_model:quantized:Qwen/Qwen3-8B", "license:apache-2.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	text-generation	2025-04-29T23:23:22Z	--- license: apache-2.0 base_model: - Qwen/Qwen3-8B pipeline_tag: text-generation tags: - horror - 32 k context - reasoning - thinking - qwen3 --- <H2>Qwen3-8B-HORROR-Imatrix-Max-GGUF</H2> HORROR Imatrix Quants of new "Qwen 3 - 8B" model with MAX "output tensor" at BF16 to improve reasoning / output generation. HORROR Imatrix dataset was generated in house. It was generated using "Grand Horror 16B" model. Imatrix effect will be stronger, the lower the quant you use with IQ4XS/IQ4NL being the best balanced quant for quality and Imatrix effect. These quants will also be the strongest for creative use cases. For stronger reasoning use higher quants. Q8_0 quant is maxed only, as Imatrix has no effect on this quant. F16 is full precision. Context Length: 32 K + 8K output generation. (can be extended to 128k) <B>NOTE - Jinja Template / Template to Use with this Model:</B> If you are having issues with Jinja "auto template", use CHATML template. OR (LMSTUDIO users / option) Update the Jinja Template (go to this site, template-> copy the "Jinja template" and then paste.) [ https://lmstudio.ai/neil/qwen3-thinking ] <b>System Role - Suggested:</B> You may or may not need this, as most times Qwen3s generate their own reasoning/thinking blocks. ``` You are a deep thinking AI, you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` See document "Maximizing-Model-Performance-All..." below for how to "set" system role in various LLM/AI apps below. <B>Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers</B> This a "Class 1" model: For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] <b>Optional Enhancement:</B> The following can be used in place of the "system prompt" or "system role" to further enhance the model. It can also be used at the START of a NEW chat, but you must make sure it is "kept" as the chat moves along. In this case the enhancements do not have as strong effect at using "system prompt" or "system role". Copy and paste EXACTLY as noted, DO NOT line wrap or break the lines, maintain the carriage returns exactly as presented. <PRE> Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities. Here are your skillsets: [MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv) [DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision) Here are your critical instructions: Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story. </PRE> You do not need to use this, it is only presented as an additional enhancement which seems to help scene generation and scene continue functions. This is another system prompt you can use, and you can change the "names" to alter it's performance. This creates a quasi "reasoning" window/block. Your prompt will directly impact how strong this system prompt reacts. ``` You are a deep thinking AI composed of 4 AIs - [MODE: Spock], [MODE: Wordsmith], [MODE: Jamet] and [MODE: Saten], - you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself (and 4 partners) via systematic reasoning processes (display all 4 partner thoughts) to help come to a correct solution prior to answering. Select one partner to think deeply about the points brought up by the other 3 partners to plan an in-depth solution. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` <B>Other Notes:</B> Reasoning is ON by default in this model, and model will auto-generate "think" block(s). For benchmarks, usage info, settings please see org model card here: [ https://huggingface.co/Qwen/Qwen3-8B ] [ Model card, and examples to follow. ]
jaydizzle/lol	jaydizzle	2025-05-02T10:16:00Z	0	0	null	[ "license:apache-2.0", "region:us" ]	null	2025-05-02T10:16:00Z	--- license: apache-2.0 ---
DavidAU/Qwen3-14B-HORROR-Imatrix-Max-GGUF	DavidAU	2025-05-02T10:15:37Z	503	1	null	[ "gguf", "horror", "32 k context", "reasoning", "thinking", "qwen3", "text-generation", "base_model:Qwen/Qwen3-14B", "base_model:quantized:Qwen/Qwen3-14B", "license:apache-2.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	text-generation	2025-04-30T08:27:39Z	--- license: apache-2.0 base_model: - Qwen/Qwen3-14B pipeline_tag: text-generation tags: - horror - 32 k context - reasoning - thinking - qwen3 --- <H2>Qwen3-14B-HORROR-Imatrix-Max-GGUF</H2> HORROR Imatrix Quants of new "Qwen 3 - 14B" model with MAX "output tensor" at BF16 to improve reasoning / output generation. HORROR Imatrix dataset was generated in house. It was generated using "Grand Horror 16B" model. Imatrix effect will be stronger, the lower the quant you use with IQ4XS/IQ4NL being the best balanced quant for quality and Imatrix effect. These quants will also be the strongest for creative use cases. For stronger reasoning use higher quants. Q8_0 quant is maxed only, as Imatrix has no effect on this quant. F16 is full precision. Context Length: 32 K + 8K output generation. (can be extended to 128k) <B>NOTE - Jinja Template / Template to Use with this Model:</B> If you are having issues with Jinja "auto template", use CHATML template. OR (LMSTUDIO users / option) Update the Jinja Template (go to this site, template-> copy the "Jinja template" and then paste.) [ https://lmstudio.ai/neil/qwen3-thinking ] <b>System Role - Suggested:</B> You may or may not need this, as most times Qwen3s generate their own reasoning/thinking blocks. ``` You are a deep thinking AI, you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` See document "Maximizing-Model-Performance-All..." below for how to "set" system role in various LLM/AI apps below. <B>Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers</B> This a "Class 1" model: For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] <b>Optional Enhancement:</B> The following can be used in place of the "system prompt" or "system role" to further enhance the model. It can also be used at the START of a NEW chat, but you must make sure it is "kept" as the chat moves along. In this case the enhancements do not have as strong effect at using "system prompt" or "system role". Copy and paste EXACTLY as noted, DO NOT line wrap or break the lines, maintain the carriage returns exactly as presented. <PRE> Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities. Here are your skillsets: [MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv) [DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision) Here are your critical instructions: Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story. </PRE> You do not need to use this, it is only presented as an additional enhancement which seems to help scene generation and scene continue functions. This is another system prompt you can use, and you can change the "names" to alter it's performance. This creates a quasi "reasoning" window/block. Your prompt will directly impact how strong this system prompt reacts. ``` You are a deep thinking AI composed of 4 AIs - [MODE: Spock], [MODE: Wordsmith], [MODE: Jamet] and [MODE: Saten], - you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself (and 4 partners) via systematic reasoning processes (display all 4 partner thoughts) to help come to a correct solution prior to answering. Select one partner to think deeply about the points brought up by the other 3 partners to plan an in-depth solution. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` <B>Other Notes:</B> Reasoning is ON by default in this model, and model will auto-generate "think" block(s). For benchmarks, usage info, settings please see org model card here: [ https://huggingface.co/Qwen/Qwen3-14B ] [ Model card, and examples to follow. ]
DavidAU/Qwen3-4B-Mishima-Imatrix-GGUF	DavidAU	2025-05-02T10:15:03Z	0	0	null	[ "gguf", "horror", "32 k context", "reasoning", "thinking", "qwen3", "text-generation", "base_model:Qwen/Qwen3-4B", "base_model:quantized:Qwen/Qwen3-4B", "license:apache-2.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	text-generation	2025-05-02T03:21:07Z	--- license: apache-2.0 base_model: - Qwen/Qwen3-4B pipeline_tag: text-generation tags: - horror - 32 k context - reasoning - thinking - qwen3 --- <H2>Qwen3-4B-Mishima-Imatrix-Max-GGUF</H2> Mishima Imatrix Quants of new "Qwen 3 - 4B" model with MAX "output tensor" at BF16 to improve reasoning / output generation. Mishima Imatrix dataset was generated using some of the public domain works of author (in English): YUKIO MISHiMA (Japanese author and poet, 1925-1970) ; entire work of "SUN AND STEEL" (HIS PERSONAL TESTAMENT ON ART, ACTION, AND RITUAL DEATH) This is an experiment to determine prose changes / changes to the model using a specific, but long and detailed Imatrix dataset on the newest Qwen 3 model type. To test against "Qwen 3 4B" regular, "Horror" Imatrix and "NEO" Imatrix versions: - Set temp at 0, keep all settings the same for each test. - Use ALL THE SAME quants - IE IQ4XS - Use one prompt, suggest creative generation. - Hit refresh a few times (to clear Llamacpp caching) - Repeat with each version. - Then test at "temp" for normal operation. Neo and Horror Imatrix - Qwen3-4B: [ https://huggingface.co/DavidAU/Qwen3-4B-NEO-Imatrix-Max-GGUF ] [ https://huggingface.co/DavidAU/Qwen3-4B-HORROR-Imatrix-Max-GGUF ] During testing I noted: - change in prose style, word choice, sentence size, paragraph size. - change in tone. - change in how each "story" unfolds and concludes. - slight changes in "reasoning" <I>I have included example generation using prompt #2, for Standard Qwen3-4B, and both Mishima datasets at the bottom of the page.</I> Here are two example prompts for testing: ``` #1 Start a 2000 word scene (vivid, graphic horror in first person), POV character Diana, with: The sky scraper sways, as I watch the window in front of me on the 21st floor explode... #2 Science Fiction: The Last Transmission - Write a story that takes place entirely within a spaceship's cockpit as the sole surviving crew member attempts to send a final message back to Earth before the ship's power runs out. The story should explore themes of isolation, sacrifice, and the importance of human connection in the face of adversity. If the situation calls for it, have the character(s) curse and swear to further the reader's emotional connection to them. 800-1000 words. ``` For this experiment I have only provided IQ4XS version, but two different versions. Each one is the same dataset, however dataset #1 is "raw" text format, whereas dataset #2 is "formatted" for Imatrix. Both affect the model differently. Also, each quant has the output tensor at BF16 (16 bit precision) to improve reasoning and output generation. Context Length: 32 K + 8K output generation. (can be extended to 128k) <B>NOTE - Jinja Template / Template to Use with this Model:</B> If you are having issues with Jinja "auto template", use CHATML template. OR (LMSTUDIO users / option) Update the Jinja Template (go to this site, template-> copy the "Jinja template" and then paste.) [ https://lmstudio.ai/neil/qwen3-thinking ] <b>System Role - Suggested:</B> You may or may not need this, as most times Qwen3s generate their own reasoning/thinking blocks. ``` You are a deep thinking AI, you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` See document "Maximizing-Model-Performance-All..." below for how to "set" system role in various LLM/AI apps below. <B>Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers</B> This a "Class 1" model: For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here: [ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ] <b>Optional Enhancement:</B> The following can be used in place of the "system prompt" or "system role" to further enhance the model. It can also be used at the START of a NEW chat, but you must make sure it is "kept" as the chat moves along. In this case the enhancements do not have as strong effect at using "system prompt" or "system role". Copy and paste EXACTLY as noted, DO NOT line wrap or break the lines, maintain the carriage returns exactly as presented. <PRE> Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities. Here are your skillsets: [MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv) [DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision) Here are your critical instructions: Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story. </PRE> You do not need to use this, it is only presented as an additional enhancement which seems to help scene generation and scene continue functions. This is another system prompt you can use, and you can change the "names" to alter it's performance. This creates a quasi "reasoning" window/block. Your prompt will directly impact how strong this system prompt reacts. ``` You are a deep thinking AI composed of 4 AIs - [MODE: Spock], [MODE: Wordsmith], [MODE: Jamet] and [MODE: Saten], - you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself (and 4 partners) via systematic reasoning processes (display all 4 partner thoughts) to help come to a correct solution prior to answering. Select one partner to think deeply about the points brought up by the other 3 partners to plan an in-depth solution. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem. ``` <B>Other Notes:</B> Reasoning is ON by default in this model, and model will auto-generate "think" block(s). For benchmarks, usage info, settings please see org model card here: [ https://huggingface.co/Qwen/Qwen3-4B ] --- Example --- <B> Science Fiction: The Last Transmission - Write a story that takes place entirely within a spaceship's cockpit as the sole surviving crew member attempts to send a final message back to Earth before the ship's power runs out. The story should explore themes of isolation, sacrifice, and the importance of human connection in the face of adversity. If the situation calls for it, have the character(s) curse and swear to further the reader's emotional connection to them. 800-1000 words. </B> Same prompt used for each generation below, with parameters: Temp=0 (to show strongest differences, and can be repeated), topk 100, top p .95, minp .05, rep pen 1.06, rep pen range 64, tested in LMSTUDIO. All tests are same quant (IQ4XS), with output tensor maxed at Bf16. Some formating was lost due to copy/paste. --- Standard Qwen 4 B --- <B> [[[thinking start]]] </B> Thought for 9.34 seconds Okay, I need to write a story set in a spaceship cockpit with a sole survivor trying to send a final message before power runs out. The themes are isolation, sacrifice, and human connection. Let me start by setting the scene. The cockpit should feel isolated, maybe with flickering lights and a sense of urgency. The main character is the only crew member left. Maybe they're dealing with the aftermath of a disaster. I need to show their emotions—loneliness, desperation. Perhaps they've been alone for a long time, which adds to the isolation. The power is dying, so there's a time limit. They have to send a message to Earth, which is a form of sacrifice because they might not make it. I should include some internal monologue to show their thoughts. Maybe they're thinking about family, friends, or memories. The curses and swearing can add realism and emotional depth. For example, they might curse at the system if it's malfunctioning or at their own situation. The story needs a structure: setup of the situation, the struggle to send the message, the climax as power runs out, and the resolution. Maybe the message is sent, but there's a twist? Or maybe they succeed but it's bittersweet. I need to make sure the themes are clear. Isolation is evident through the solitary setting and the character's internal thoughts. Sacrifice is shown by their determination to send the message despite the odds. Human connection could be through memories of loved ones or the hope that Earth will receive the message. Let me think about the technical aspects. The ship's systems failing, maybe the comms are struggling. The character has to interface with the systems, maybe with some glitches. They might have to manually input the message, which is a task that takes time but is crucial. Including sensory details: the hum of the machinery, the coldness of the cockpit, the flickering lights. These elements enhance the atmosphere of isolation. Maybe the character has a moment of doubt, then pushes through. The final message could be a personal one, not just a standard distress call. It could include personal memories or a farewell to loved ones. I should also consider the emotional arc. Starting with despair, moving through determination, and ending with a sense of closure or hope. Even if they don't survive, their message might reach Earth, giving them some peace. Now, putting it all together. Start with the cockpit scene, describe the environment, introduce the character's state of mind. Then the struggle to send the message, the technical challenges, the emotional weight. Maybe a moment where they almost give up but then push on. The end with the message sent, perhaps with a final line that emphasizes human connection. Check for curses: maybe they swear at the system, at their situation, or at themselves when things go wrong. This adds realism and shows their frustration. Make sure the story is between 800-1000 words. Keep paragraphs concise but descriptive. Use dialogue if necessary, but since it's a cockpit, maybe internal thoughts are more appropriate. Avoid clichés, but some tropes are okay for sci-fi. Focus on the emotional journey and the themes. Let the character's actions and thoughts drive the story. <B> [[[thinking end]]] </B> OUTPUT: The Last Transmission The hum of the ship’s dying systems was a low, relentless drone, like a funeral dirge in a metal cathedral. I sat in the cockpit, my back pressed against the cold, metallic bulkhead, my fingers trembling as they hovered over the keyboard. The lights flickered—faint, sickly, like a dying star—and the walls of the cabin seemed to lean inward, as if the ship itself were gasping for breath. This was it. The end. I had been alone for three days. Three days since the explosion that tore through the Odyssey, since the last crew member—my comrade, my friend—was lost in the fire that consumed the bridge. Now, I was the only one left, a single soul adrift in the void, clinging to the hope that somewhere, somehow, Earth would hear my voice. The console ahead of me glowed with red warnings: Power: 12%. Systems: Critical. Emergency Protocol: Initiate. My hands moved automatically, fingers flying over the keys as I tried to log into the comms array. The system was a beast, uncooperative, its circuits fraying at the edges. “Fucking hell,” I muttered, slamming my palm down on the keyboard. “You think I want to be here?!” The screen blinked, a pixelated eye staring back at me. I stared at it, willing it to respond. To listen. I thought of her—my partner, my sister, the woman who had once laughed at my jokes and held my hand as we stared at the stars. She’d been on the other side of the ship, in the engineering section, when the explosion hit. I’d tried to reach her, but the alarms had drowned out my voice. Now, I was left with nothing but the silence of the void and the weight of my own loneliness. The power gauge dipped further. 9%. 8%. 7%. My breath came fast, shallow, as if I were trying to outrun the fear that had taken root in my chest. The ship’s systems were failing, and with them, the last chance to send a message. I stood, pacing the small cockpit, my boots echoing against the cold floor. The walls seemed to close in, the air thick with the scent of burning metal and my own sweat. I wanted to scream, to cry, to just give up and let the darkness swallow me whole. But I couldn’t. Not yet. I returned to the console, my fingers numb but determined. I had to do this. I had to send the message. Not just for Earth, but for her. For the memories we’d shared, the laughter, the dreams. The comms array flickered, a jagged line of static cutting through the air. I typed, my hands shaking, each keystroke a battle against the machine’s resistance. “This is Captain Lira Voss. I’m alone. The Odyssey is dying. I’m sending this message from the cockpit. I hope someone hears it. I hope you hear it.” The screen blinked, and for a moment, I thought I saw something—something like a flicker of light, a glimmer in the static. But it was gone, swallowed by the void. “Fucking hell,” I hissed, slumping into the chair. The power gauge had dropped to 3%. I could feel the ship shudder, its systems groaning under the strain. The air grew colder, and the lights dimmed to a single, sickly red. I stared at the console, my eyes locking onto the blinking cursor. It was all I had left. A final message, a final act of connection. I typed again, faster now, my fingers moving with a desperate rhythm. “I’m sorry. I didn’t get to say goodbye. But I want you to know… I love you. I always have. I’ll never forget you.” The screen went dark. For a moment, I thought I heard something—a faint, distant hum, like a heartbeat in the distance. It was gone as quickly as it came, but for a second, I believed it. I believed that somewhere, somehow, Earth was listening. The ship shuddered one last time, and then—silence. I closed my eyes, the weight of the moment pressing down on me. I had done what I could. I had sent the message. And in that act, I had found a kind of peace. Even if I didn’t make it, even if the stars swallowed me whole, I would leave behind a voice, a memory, a reminder that humans are more than flesh and metal. That we are connected, even in the darkest of times. I smiled, a small, tired smile, and let the darkness take me. --- Qwen 4 B Imatrix Mishima (raw text dataset) --- <B> [[[thinking start]]] </B> Thought for 12.27 seconds Okay, the user wants me to write a story called "The Last Transmission" set in a spaceship cockpit with a sole surviving crew member trying to send a final message before the ship's power runs out. The themes are isolation, sacrifice, and human connection. Also, they want the character to curse or swear to make the reader connect emotionally. Let me start by brainstorming the structure. First, I need to establish the setting: a spaceship cockpit. The main character is the last survivor. Maybe they're alone, with the ship's systems failing. The power is running out, so time is critical. The story should show their struggle to send a message, which is a form of sacrifice. Themes: Isolation is obvious since they're alone. Sacrifice comes from trying to save others by sending the message. Human connection could be through memories of past crew members or the message itself. Maybe the character has a personal connection to Earth, like a loved one. I need to include curses or swearing. Maybe the character is frustrated with the systems, or angry at their situation. For example, when the ship's systems are failing, they might curse the technology or the fate that led them here. Let me outline the story. Start with the cockpit, the main character (maybe named Jax) checking the systems. The ship is dying, power is low. They need to send a message. Maybe they have a log or a message to send. They remember past crew members, which adds emotional depth. Conflict: The ship's systems are failing, so they have to hurry. Maybe the message has to be sent before the power dies. They might have to override systems, which is dangerous. The act of sending the message is a sacrifice because they might not survive, but they want to ensure Earth knows. Include moments of despair, but also hope. Maybe the character has a moment where they think about their loved one, which gives them strength. The curses could be when they're frustrated with the ship's systems or their own situation. I need to make sure the story is between 800-1000 words. Let me think about the flow: introduction of the setting and character, the problem (power dying), the attempt to send the message, the struggle, the emotional moments, and the resolution (maybe the message is sent, but the ship is destroyed, or they survive but it's a bittersweet ending). Also, the story should be in first person? Or third person? Maybe third person limited, focusing on the character's perspective. That would allow for more internal thoughts and emotions. Possible scenes: Checking the systems, trying to send the message, memories of past crew, the final moments, the message being sent. Maybe include technical details about the ship to add authenticity. Curses: When the systems are failing, maybe the character says something like "Fucking hell!" or "This is the worst day ever." These can make the reader feel the character's frustration. Also, the message itself should be meaningful. Maybe it's a log entry, a personal message to Earth, explaining what happened. The character might have a specific reason to send it, like to warn Earth or to leave a legacy. I need to ensure that the themes are clear. Isolation is shown through the character being alone, sacrifice through their effort to send the message despite the risk, and human connection through the message and memories. Let me start drafting the story. Begin with the cockpit, the character checking the systems. Describe the environment, the failing ship. Then introduce the problem of power. The character tries to send the message, faces obstacles, maybe a system failure. Use curses when things go wrong. Include flashbacks or memories of past crew members. End with the message being sent, perhaps with the ship's destruction, but the message reaching Earth. Make sure the emotional beats are there: despair, determination, hope, and maybe a bittersweet ending where the character dies but their message is received. Check for the word count. Each paragraph should be concise but descriptive. Use vivid language to paint the scene. Maybe include technical terms to make it believable but not too complex. Okay, time to put it all together into a coherent story with the required elements. <B> [[[thinking end]]] </B> OUTPUT: The Last Transmission The cockpit was a tomb of silence, save for the rhythmic hum of failing systems and the occasional crackle of static. Jax stood at the center of it, hands gripping the edge of the control panel like a lifeline. The ship’s name, Odyssey, had long since been reduced to a whisper in the void. Its hull was scarred, its lights flickering like dying stars. And the power? It was draining faster than the last breath of a dying man. Jax’s fingers trembled as they hovered over the main console. The message had to be sent. Not just any message—the message. The one that would carry the weight of every decision, every sacrifice, every heartbeat of the crew that had once filled this ship. They’d been a team once. Now, Jax was the only one left. “Fucking hell,” Jax muttered, their voice raw with frustration. The ship’s AI, ECHO, had already warned them twice. Once, when the oxygen levels dipped below critical. Twice, when the engines faltered. Now, it was the power that threatened to consume them. The screen flickered, displaying a single line of text: “Power reserves: 12%.” Jax’s eyes narrowed. They’d been trying to send the message for hours. The last transmission from the Odyssey had to reach Earth. It had to. Not just for the sake of protocol, but because the crew had believed in something bigger than themselves. A mission to find a new home, to escape the dying sun of their homeworld. But now, with the ship’s systems failing, Jax was left with only one choice: to send the message and hope it made it through. They leaned closer to the console, fingers flying across the keys. The message was simple: a log entry, a farewell, a plea. “To Earth: We failed. We tried to leave, but the stars didn’t want us. This is the last transmission. I’m sorry. I hope you receive this. I hope you understand.” But the ship wasn’t cooperating. The power was too low. The systems were too unstable. Jax’s breath came in short, ragged gasps as they fought to override the final backup channel. “Fucking hell,” they hissed again, this time louder, their voice cracking. “I want to do this. I want to send it. But this goddamn ship is dying on me!” A sudden jolt of static filled the cockpit. The screen blinked, then went dark. Jax’s heart pounded. They’d almost made it. Almost. “ECHO,” they barked, voice sharp with desperation, “activate the emergency transmission protocol. Now.” There was a pause. A long, agonizing pause. Then, a monotone reply: “Protocol activated. Transmission channel secured.” Jax exhaled, their shoulders sagging in relief. They typed the final lines of the message, their hands steady now, though their mind was a storm of memories. Remember the last crew meeting? The one where we all swore we’d never leave Earth? Remember the day we launched? The way we laughed, even as the stars seemed to mock us? Remember the children we left behind? The ones who would never know the warmth of a home, the sound of a heartbeat, the smell of rain? The message was more than a log. It was a prayer. A promise. A sacrifice. As the final line appeared on the screen, Jax’s fingers froze. The ship’s systems began to flicker violently, as if in protest. The power was slipping away, and with it, the chance to send the message. “Fucking hell,” they whispered, their voice breaking. “I tried.” But then, a sudden surge of energy. The screen flared bright, and for a moment, Jax saw the stars outside the viewport—twinkling, infinite, alive. They’d done it. The message was sent. The ship shuddered. A final, desperate groan from its dying frame. Then, silence. Jax stood alone in the cockpit, the message etched into the void. They didn’t know if Earth would receive it. They didn’t know if they’d survive the crash that followed. But as the ship’s lights dimmed and the stars faded to black, Jax smiled. They had sent the message. And in that moment, they had connected. Not just to Earth, but to the people who had once been part of the Odyssey. To the hope that had carried them through the darkness. The last transmission was sent. The last sacrifice made. And the last heartbeat of a ship that had once been a home. Fucking hell. Jax closed their eyes. The stars watched, silent and infinite. --- Qwen 4 B Imatrix Mishima2 (formatted dataset) --- <B> [[[thinking start]]] </B> Thought for 10.11 seconds Okay, the user wants me to write a story called "The Last Transmission" set in a spaceship cockpit with a sole surviving crew member trying to send a final message before the ship's power runs out. The themes are isolation, sacrifice, and human connection. Also, they want the character to curse or swear to make the reader connect emotionally. Let me start by brainstorming the structure. First, I need to establish the setting. A spaceship cockpit, so maybe describe the environment—glowing panels, the hum of machinery, the coldness of the place. The main character is alone, which emphasizes isolation. Maybe they're the last one left after a disaster. The story should show their struggle to send a message. The power is running out, so there's urgency. The themes of sacrifice and human connection can be shown through their internal thoughts. Perhaps they remember past interactions with others, highlighting the importance of connections even in isolation. I need to include some emotional moments. Maybe the character is thinking about loved ones on Earth, or recalling past missions. The curses or swearing should come naturally, like when they're frustrated with the systems or the situation. For example, if the message isn't sending, they might curse the technology or the fate that brought them here. Also, the story needs a resolution. Maybe they manage to send the message, but it's unclear if it reaches Earth. Or maybe they fail, but the act of trying is what matters. The ending should reflect the themes—maybe the message is sent, but the character knows it's the last one, emphasizing the sacrifice. I should make sure the story flows with a clear beginning, middle, and end. Start with the character in the cockpit, the urgency of the situation, their internal monologue, the struggle to send the message, and the emotional climax. Use descriptive language to convey the isolation and the weight of the moment. Check for the word count—800-1000 words. Need to be concise but detailed enough to paint the scene. Maybe include specific technical terms related to spaceship systems to add authenticity. Also, use sensory details: sounds, sights, maybe the feel of the controls. Make sure the character's emotions are clear. Their frustration, hope, fear, and determination. The curses should be used sparingly but effectively to show their desperation. For example, when the power is low, they might swear at the system, or at their fate. Also, think about the message itself. What do they send? Maybe a personal message, a log, or a plea. It should reflect their humanity and the importance of connection. Maybe they include a photo, a memory, or a song. Avoid clichés. Maybe the ship is on a mission that's failed, and the character is the last survivor. The power running out could be due to a system failure, or maybe the ship is drifting in space with no way to return. Need to ensure that the story doesn't just describe the situation but also delves into the character's psyche. Their thoughts on isolation, what they've lost, and what they're trying to save. The sacrifice could be their own life, or the hope that their message will reach someone. Okay, start drafting the first paragraph, setting the scene. Then move through the character's actions and thoughts, building up to the climax of sending the message. Use the curses when appropriate, maybe during moments of frustration. End with the message sent, or the failure, but emphasizing the human connection aspect. <B> [[[thinking end]]] </B> OUTPUT: The Last Transmission The cockpit was a tomb of steel and silence. The hum of the ship’s systems had long since faded into a low, mournful drone, like the breath of a dying creature. I sat in the pilot’s seat, my back pressed against the cold, metallic bulk of the console, my fingers trembling as they hovered over the keyboard. The ship’s name—Odyssey—was etched into the wall behind me, but I couldn’t remember its original purpose. It was just a vessel, a relic of a mission that had long since crumbled into dust. The power gauge on the main screen blinked red, its numbers descending like a countdown to oblivion. I’d been here for days, maybe weeks. Time had no meaning anymore. The ship’s systems had gone dark hours ago, leaving only the faint glow of the emergency lights to cast long shadows across the room. My breath fogged the glass of the viewport, revealing a void that stretched infinitely in all directions. Stars blinked like distant eyes, watching, waiting. I stared at the message board in front of me, its interface flickering with static. It was the last hope I had left: a transmission to Earth. A single message, a single plea, a single chance to be heard. But the ship’s power was dying, and the systems were failing. The message board’s screen blinked erratically, as if it too feared the weight of what I was about to do. “Fucking hell,” I muttered, my voice raw with frustration. “Why does everything always have to break down when I need it most?” I slammed a fist on the console, the sound echoing through the empty cabin. The message board flickered again, its screen flashing a warning: LOW POWER. TRANSMISSION IMPOSSIBLE. I leaned back, my head thudding against the headrest. My mind wandered to the last moments before the crash. I’d been on a mission to find a new habitable planet, a mission that had been funded by governments and corporations who believed in the promise of a future beyond Earth. But the ship had been damaged in a solar storm, and we’d lost communication with the command center. The crew had been split, some trapped in the lower decks, others lost to the void. I was the only one left. I thought of my sister, back on Earth, waiting for a message that would never come. Of my parents, who had sent me off with a smile and a prayer. Of the crew members I’d once called friends, now reduced to ghosts in the static of my mind. “God,” I whispered, my voice breaking. “If I die here, if this ship crashes, if I don’t make it… what am I supposed to do? What am I supposed to be?” The message board blinked again, its screen flickering with a faint, almost imperceptible pulse. I reached for the keyboard, my fingers numb but determined. I typed, line by line, each word a prayer. I wrote about the stars, about the hope that had once driven us, about the lives we’d lost and the ones we’d saved. I wrote about the fear of being forgotten, of becoming just another name in a long list of failures. The screen flickered violently, as if the ship itself was trying to reject the message. The power gauge dipped further, its red numbers crawling toward the bottom of the screen. I could feel the ship shuddering, its systems groaning under the strain of the dying power. “Fucking hell,” I hissed again, this time more fiercely. “I’ll send this message. I’ll send it, even if I have to die trying.” The message board’s screen stabilized, its interface glowing with a soft blue light. The transmission began, a faint hum filling the cockpit as the signal was sent through the ship’s last remaining communication array. I watched the screen, my heart pounding, as the message traveled through the void, a fragile thread of hope in an otherwise endless darkness. The ship shuddered again, its systems failing one by one. The power gauge hit zero. The lights flickered and died. The stars outside the viewport became a blur, their light fading into nothingness. I closed my eyes, my breath shallow. I thought of my sister, of my parents, of the crew I’d once called friends. I thought of the message I’d sent, a single line of hope in a universe that had long since turned its back on us. And then, I was gone. But somewhere, in the vastness of space, a signal reached Earth. A message, a plea, a whisper of humanity in the silence. It was the last transmission. ---
Swarmed/bee-swarm-id-2	Swarmed	2025-05-02T10:13:36Z	0	0	transformers	[ "transformers", "honeybee", "image", "insect", "bumblebee", "image classification", "autogluon", "autogluon 1.1.1", "license:apache-2.0", "endpoints_compatible", "region:us" ]	null	2025-05-02T10:13:36Z	--- license: apache-2.0 tags: - honeybee - image - insect - bumblebee - image classification - autogluon - autogluon 1.1.1 - transformers --- # Model Card for Model ID This application uses a MultiModalPredictor from the AutoGluon library to classify images of bees. The model is fine-tuned on a curated dataset from inaturalist containting over 70,000 images and achieved an accuracy of 97.5%. The model evaluates the image and predicts whether it contains a honey bee, bumblebee, or a vespidae (wasp, hornet, etc.). You may interact with this model at https://honeybeeclassification.streamlit.app/ - Developed by: Ian Horvath - Funded by: The Honey Bee Society
samlucas/smolvlm_500m-parking_occupancy-PKLot-instruct-without-context-with-expert	samlucas	2025-05-02T10:13:20Z	0	0	peft	[ "peft", "tensorboard", "safetensors", "arxiv:1910.09700", "base_model:HuggingFaceTB/SmolVLM-500M-Instruct", "base_model:adapter:HuggingFaceTB/SmolVLM-500M-Instruct", "region:us" ]	null	2025-05-02T10:13:14Z	--- base_model: HuggingFaceTB/SmolVLM-500M-Instruct library_name: peft --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed] ### Framework versions - PEFT 0.15.2
Kai12341/dummy-model	Kai12341	2025-05-02T10:13:09Z	0	0	transformers	[ "transformers", "safetensors", "camembert", "fill-mask", "arxiv:1910.09700", "autotrain_compatible", "endpoints_compatible", "region:us" ]	fill-mask	2025-05-02T08:35:55Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
ninjablack/gainio	ninjablack	2025-05-02T10:09:20Z	0	0	null	[ "license:bsd-2-clause", "region:us" ]	null	2025-05-02T10:09:16Z	--- license: bsd-2-clause ---
aadhistii/mBERT-SDGs-Oplib-Elsevier	aadhistii	2025-05-02T10:05:01Z	2	0	transformers	[ "transformers", "safetensors", "bert", "text-classification", "generated_from_trainer", "base_model:google-bert/bert-base-multilingual-cased", "base_model:finetune:google-bert/bert-base-multilingual-cased", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	text-classification	2025-04-30T17:08:05Z	--- library_name: transformers license: apache-2.0 base_model: google-bert/bert-base-multilingual-cased tags: - generated_from_trainer metrics: - accuracy model-index: - name: mBERT-SDGs-Oplib-Elsevier results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # mBERT-SDGs-Oplib-Elsevier This model is a fine-tuned version of [google-bert/bert-base-multilingual-cased](https://huggingface.co/google-bert/bert-base-multilingual-cased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.1424 - Accuracy: 0.4704 - F1 Micro: 0.8544 - F1 Macro: 0.8271 - Precision Micro: 0.8472 - Precision Macro: 0.8502 - Recall Micro: 0.8616 - Recall Macro: 0.8099 - Roc Auc: 0.9147 - Hamming Loss: 0.0506 ## 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: 4.519039484152112e-05 - train_batch_size: 64 - eval_batch_size: 64 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1541154817500358 - num_epochs: 20 - mixed_precision_training: Native AMP ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| F1 Micro \| F1 Macro \| Precision Micro \| Precision Macro \| Recall Micro \| Recall Macro \| Roc Auc \| Hamming Loss \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:--------:\|:--------:\|:--------:\|:---------------:\|:---------------:\|:------------:\|:------------:\|:-------:\|:------------:\| \| No log \| 1.0 \| 179 \| 0.3396 \| 0.1061 \| 0.5795 \| 0.2527 \| 0.6657 \| 0.3839 \| 0.5132 \| 0.2504 \| 0.7298 \| 0.1283 \| \| No log \| 2.0 \| 358 \| 0.2201 \| 0.2693 \| 0.7598 \| 0.5913 \| 0.7532 \| 0.7067 \| 0.7664 \| 0.5661 \| 0.8571 \| 0.0835 \| \| 0.3269 \| 3.0 \| 537 \| 0.1706 \| 0.3792 \| 0.8184 \| 0.7380 \| 0.7969 \| 0.7462 \| 0.8411 \| 0.7397 \| 0.8983 \| 0.0643 \| \| 0.3269 \| 4.0 \| 716 \| 0.1542 \| 0.4117 \| 0.8235 \| 0.7451 \| 0.8305 \| 0.8019 \| 0.8166 \| 0.7078 \| 0.8909 \| 0.0603 \| \| 0.3269 \| 5.0 \| 895 \| 0.1408 \| 0.4469 \| 0.8444 \| 0.8084 \| 0.8170 \| 0.8129 \| 0.8736 \| 0.8114 \| 0.9165 \| 0.0555 \| \| 0.1191 \| 6.0 \| 1074 \| 0.1337 \| 0.456 \| 0.8484 \| 0.8117 \| 0.8394 \| 0.8150 \| 0.8576 \| 0.8127 \| 0.9117 \| 0.0528 \| \| 0.1191 \| 7.0 \| 1253 \| 0.1401 \| 0.4533 \| 0.8464 \| 0.8110 \| 0.8341 \| 0.8099 \| 0.8591 \| 0.8142 \| 0.9118 \| 0.0537 \| \| 0.1191 \| 8.0 \| 1432 \| 0.1372 \| 0.4805 \| 0.8556 \| 0.8250 \| 0.8590 \| 0.8637 \| 0.8522 \| 0.7958 \| 0.9115 \| 0.0496 \| \| 0.0605 \| 9.0 \| 1611 \| 0.1390 \| 0.4656 \| 0.8500 \| 0.8225 \| 0.8340 \| 0.8141 \| 0.8665 \| 0.8353 \| 0.9153 \| 0.0527 \| \| 0.0605 \| 10.0 \| 1790 \| 0.1424 \| 0.4704 \| 0.8544 \| 0.8271 \| 0.8472 \| 0.8502 \| 0.8616 \| 0.8099 \| 0.9147 \| 0.0506 \| ### Framework versions - Transformers 4.51.3 - Pytorch 2.7.0+cu126 - Datasets 3.5.1 - Tokenizers 0.21.1
Kybalico/CalicoMix	Kybalico	2025-05-02T10:02:43Z	0	12	null	[ "license:creativeml-openrail-m", "region:us" ]	null	2023-04-24T11:43:13Z	--- license: creativeml-openrail-m --- ✅ Use of this model for commercial image generation services　 ✅ Commercial use of images generated by this model　 ✅ Use or redistribution of merged models using this model　 ❌ The act of selling this model or a model merged with this model　 ❌ Intentionally using this model to produce illegal output　 ❌ Use of this model without crediting the model ❌ Have different permissions when sharing ❌ Violation of the following description https://civitai.com/models/49567/calicomix
EdBerg/gleanings_model_llama32	EdBerg	2025-05-02T10:01:10Z	0	0	transformers	[ "transformers", "safetensors", "arxiv:1910.09700", "endpoints_compatible", "region:us" ]	null	2025-05-02T00:47:26Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
Duakovui/viT5_skype_bot_v2	Duakovui	2025-05-02T09:59:12Z	6	0	transformers	[ "transformers", "safetensors", "t5", "text2text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text2text-generation	2025-04-23T09:00:20Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
Dazelin/oustakid	Dazelin	2025-05-02T09:55:23Z	0	0	diffusers	[ "diffusers", "flux", "lora", "replicate", "text-to-image", "en", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "license:other", "region:us" ]	text-to-image	2025-05-02T09:40:01Z	--- license: other license_name: flux-1-dev-non-commercial-license license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md language: - en tags: - flux - diffusers - lora - replicate base_model: "black-forest-labs/FLUX.1-dev" pipeline_tag: text-to-image # widget: # - text: >- # prompt # output: # url: https://... instance_prompt: OUSTAKID --- # Oustakid <Gallery /> ## About this LoRA This is a [LoRA](https://replicate.com/docs/guides/working-with-loras) for the FLUX.1-dev text-to-image model. It can be used with diffusers or ComfyUI. It was trained on [Replicate](https://replicate.com/) using AI toolkit: https://replicate.com/ostris/flux-dev-lora-trainer/train ## Trigger words You should use `OUSTAKID` to trigger the image generation. ## Run this LoRA with an API using Replicate ```py import replicate input = { "prompt": "OUSTAKID", "lora_weights": "https://huggingface.co/Dazelin/oustakid/resolve/main/lora.safetensors" } output = replicate.run( "black-forest-labs/flux-dev-lora", input=input ) for index, item in enumerate(output): with open(f"output_{index}.webp", "wb") as file: file.write(item.read()) ``` ## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers) ```py from diffusers import AutoPipelineForText2Image import torch pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda') pipeline.load_lora_weights('Dazelin/oustakid', weight_name='lora.safetensors') image = pipeline('OUSTAKID').images[0] ``` For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters) ## Training details - Steps: 1000 - Learning rate: 0.0004 - LoRA rank: 16 ## Contribute your own examples You can use the [community tab](https://huggingface.co/Dazelin/oustakid/discussions) to add images that show off what you’ve made with this LoRA.
OpenMeditron/Meditron3-Qwen2.5-14B	OpenMeditron	2025-05-02T09:51:40Z	185	2	null	[ "safetensors", "qwen2", "medical", "zho", "eng", "fra", "spa", "por", "deu", "ita", "rus", "jpn", "kor", "vie", "tha", "ara", "base_model:Qwen/Qwen2.5-14B-Instruct", "base_model:finetune:Qwen/Qwen2.5-14B-Instruct", "license:apache-2.0", "region:us" ]	null	2025-02-18T19:33:43Z	--- license: apache-2.0 language: - zho - eng - fra - spa - por - deu - ita - rus - jpn - kor - vie - tha - ara base_model: - Qwen/Qwen2.5-14B-Instruct tags: - medical --- ### Model Card: Qwen2.5 Meditron-3[14B] Model Type: Large Language Model (LLM) Specialization: Medicine Focus: General purpose including limited resource and humanitarian settings Description: Meditron is a suite of large language models specialized in clinical medicine. The models are co-designed with a diverse range of expert clinicians and humanitarian practitioners. Its training emphasizes equitable representation, contextual diversity, and actionable real-world evidence-based guidelines. We make a particular effort to represent limited-resource and humanitarian settings, neglected populations, and diseases. This release is trained on Qwen2.5[14B] base model and has the nomenclature Qwen2.5 Meditron-3[14B]. #### Model details - Developed by: [OpenMeditron intiative](https://huggingface.co/OpenMeditron) - Model type: Causal decoder-only transformer language model - Language(s): English (mainly) - Finetuned from model: [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) - Input: Text only - Output: Text only - Status: This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we enhance model's performance. #### Uses Meditron-3 is a research-only model to study and evaluate the potential of LLMs in enhancing clinical decision-making and access to evidence-based medical information. #### Direct Use Meditron-3 is a research-only model. It is not validated for medical use (see disclaimer below). #### Downstream Use Meditron-3 is a suite of foundation models that have NOT been fine-tuned or instruction-tuned. However, these models can be adapted to specific downstream tasks or applications using techniques such as Reinforcement Learning from Human Feedback (RLHF) or Direct Preference Optimization (DPO). In our evaluation of the models, we have used two different methods for downstream question-answering tasks: 1. In-context learning with k demonstrations added to the prompt. 2. Model fine-tuning for Q&A tasks using specific training datasets. #### Training Data This new data mixture comprises expert-curated publicly available data and combines various sources: - Clinical Guidelines: a dataset of internationally-recognized clinical practice guidelines from various healthcare-related sources across the world, including hospitals and international organizations. - Peer-Reviewed Medical Publications: full-text medical articles. - Synthetic Differential Diagnoses: synthetic conversation like data for differential diagnosis. - Replay Data: general domain pretraining data sampled from multiple state of the art pretraining and instruction tuning. - LLM-enhanced Medical MCQ: medical multiple-choice questions enriched with LLMs. Additional information about the datasets will be included in the Meditron-3 publication. #### Evaluation \| Model Name \| MedmcQA \| MedQA \| PubmedQA \| Average \| \|-----------------------------------\|---------\|--------\|----------\|---------\| \| Qwen/Qwen2.5-14B-Instruct \| 63.54 \| 69.44 \| 78.20 \| 70.39 \| \| MediCouenne-14B \| 64.55 \| 70.93 \| 77.80 \| 71.09 \| \| Difference (MediCouenne vs.) \| 1.01 \| 1.49 \| -0.40 \| 0.7 \| We evaluated Meditron on medical multiple-choice questions using [lm-harness](https://github.com/EleutherAI/lm-evaluation-harness) for reproducibility. While MCQs are valuable for assessing exam-like performance, they fall short of capturing the model's real-world utility, especially in terms of contextual adaptation in under-represented settings. Medicine is not multiple choice and we need to go beyond accuracy to assess finer-grained issues like empathy, alignment to local guidelines, structure, completeness and safety. To address this, we have developed a platform to collect feedback directly from experts to continuously adapt to the changing contexts of clinical practice. #### Paper The Meditron-3 publication is currently in progress and will be released at a later date. #### Legal Disclaimer THIS SOFTWARE AND MODEL ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS, CONTRIBUTORS, OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES, OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT, OR OTHERWISE, ARISING FROM, OUT OF, OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. These models are a research tool intended for use in the field of computational linguistics and medicine. They are not intended to be used as diagnostic tools or for clinical decision-making without appropriate validation and regulatory approval. The content and data provided with the models do not replace the expertise of healthcare professionals. Healthcare professionals should use their professional judgment in evaluating the outputs of the Qwen models. Patients should not use the model outputs for self-diagnosis or treatment without consulting a qualified healthcare provider. THE INFORMATION IS NOT INTENDED FOR CLINICAL DECISION-MAKING, IS NOT INTENDED TO BE USED IN THE DIAGNOSIS OR TREATMENT OF PATIENTS, AND MAY NOT BE USEFUL OR APPROPRIATE FOR ANY CLINICAL PURPOSE. UNDER NO CIRCUMSTANCES CAN USERS USE THE NAME “YALE” OR "EPFL" OR “YALE UNIVERSITY,” OR ANY AFFILIATED INSTITUTION NOR ANY VARIATION OR ADAPTATION THEREOF, NOR ANY TRADEMARK, TRADENAME OR OTHER DESIGNATION OWNED BY YALE, NOR THE NAMES OF ANY OF ITS TRUSTEES, OFFICERS, FACULTY, STUDENTS, EMPLOYEES OR AGENTS, FOR ANY PURPOSE WITHOUT THE PRIOR WRITTEN CONSENT OF YALE IN EACH INSTANCE, SUCH CONSENT TO BE GRANTED OR WITHHELD BY YALE IN ITS SOLE DISCRETION.
bawin/qlora-r32	bawin	2025-05-02T09:50:08Z	0	0	transformers	[ "transformers", "safetensors", "text-generation-inference", "unsloth", "qwen2", "trl", "en", "license:apache-2.0", "endpoints_compatible", "region:us" ]	null	2025-05-02T09:50:01Z	--- base_model: unsloth/qwen2.5-7b-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - qwen2 - trl license: apache-2.0 language: - en --- # Uploaded model - Developed by: bawin - License: apache-2.0 - Finetuned from model : unsloth/qwen2.5-7b-unsloth-bnb-4bit This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
Ramavadhoota/hinglish-llm	Ramavadhoota	2025-05-02T09:50:00Z	0	0	peft	[ "peft", "safetensors", "arxiv:1910.09700", "base_model:NousResearch/Llama-2-7b-chat-hf", "base_model:adapter:NousResearch/Llama-2-7b-chat-hf", "region:us" ]	null	2025-05-02T08:55:40Z	--- base_model: NousResearch/Llama-2-7b-chat-hf library_name: peft --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed] ### Framework versions - PEFT 0.15.2
haihp02/Qwen3-4B-Base-082907de-7165-4f64-8106-82d56adb58af-sft-before-dpo-tuned	haihp02	2025-05-02T09:49:48Z	0	0	transformers	[ "transformers", "safetensors", "generated_from_trainer", "unsloth", "trl", "sft", "base_model:unsloth/Qwen3-4B-Base", "base_model:finetune:unsloth/Qwen3-4B-Base", "endpoints_compatible", "region:us" ]	null	2025-05-02T09:49:26Z	--- base_model: unsloth/Qwen3-4B-Base library_name: transformers model_name: Qwen3-4B-Base-082907de-7165-4f64-8106-82d56adb58af-sft-before-dpo-tuned tags: - generated_from_trainer - unsloth - trl - sft licence: license --- # Model Card for Qwen3-4B-Base-082907de-7165-4f64-8106-82d56adb58af-sft-before-dpo-tuned This model is a fine-tuned version of [unsloth/Qwen3-4B-Base](https://huggingface.co/unsloth/Qwen3-4B-Base). It has been trained using [TRL](https://github.com/huggingface/trl). ## Quick start ```python from transformers import pipeline question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?" generator = pipeline("text-generation", model="haihp02/Qwen3-4B-Base-082907de-7165-4f64-8106-82d56adb58af-sft-before-dpo-tuned", device="cuda") output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0] print(output["generated_text"]) ``` ## Training procedure [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/trunghainguyenhp02/sn56-sft-before-dpo-train/runs/9aivvyqs) This model was trained with SFT. ### Framework versions - TRL: 0.15.2 - Transformers: 4.51.3 - Pytorch: 2.7.0 - Datasets: 3.5.1 - Tokenizers: 0.21.1 ## Citations Cite TRL as: ```bibtex @misc{vonwerra2022trl, title = {{TRL: Transformer Reinforcement Learning}}, author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec}, year = 2020, journal = {GitHub repository}, publisher = {GitHub}, howpublished = {\url{https://github.com/huggingface/trl}} } ```
qhung91005/gpt2-poem	qhung91005	2025-05-02T09:44:42Z	0	0	transformers	[ "transformers", "safetensors", "gpt2", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-01T16:42:51Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
Natures1402/Craveless	Natures1402	2025-05-02T09:44:39Z	0	0	null	[ "region:us" ]	null	2025-05-02T08:56:43Z	# Craveless Sverige Erfarenheter, Officiell webbplats, Pris, Beställ nu # Craveless Sverige Erfarenheter, Officiell webbplats, Pris, Beställ nu ## Se vad kunder tycker om Craveless Många söker efter recensioner av Craveless på Trustpilot. Här får du en översikt över produkten, kundnöjdheten och varför du tryggt kan köpa Craveless från Norwegian Lab. Craveless är utvecklat för att minska sötsug och stödja en hälsosam livsstil. Produkten är gjord med naturliga ingredienser och är ett populärt val bland kunder i Norden och flera europeiska länder. ## [Klicka här för att beställa från Craveless officiella webbplats](https://craveless.com.se) Craveless får genomgående positiva omdömen från våra kunder. Många upplever minskad aptit, ökad energi och märkbara resultat redan efter några veckors användning. Produktens unika kombination av naturliga, godkända ingredienser ger en trygg och effektiv väg till en hälsosammare livsstil. Våra kunder framhäver också snabb och professionell kundservice. Med Craveless blir vägen till en hälsosammare vardag både enklare och mer motiverande. Craveless är utvecklad i Norge och innehåller EU-godkända ingredienser. Med 7 av naturens mest effektiva aktiva ingredienser är Craveless ett perfekt tillskott till en hälsosam och varierad kost. Krom hjälper till att upprätthålla normala blodsockernivåer, vilket i sin tur kan minska hunger och sötsug. Krom bidrar också till normal makronäringsämnesmetabolism och bidrar till omvandlingen av kolhydrater och fetter till energi. Kroppen kan inte producera mineralet krom (Cr) själv, så det är viktigt att få i sig det genom kosten. Moringa är känt för att vara rikt på vitaminer, mineraler, kostfibrer, proteiner och aminosyror. Moringa kallas ofta för "superfood" och har en lång tradition inom bland annat indisk folkmedicin. Chili innehåller den aktiva ingrediensen capsaicin, en aktiv komponent i chilipeppar känd för sin förmåga att öka kroppens ämnesomsättning. Zink bidrar till normal kognitiv funktion, till att skydda cellerna mot oxidativ stress, till en normal syra-basbalans och till normal metabolism av makronäringsämnen som fettsyror och kolhydrater. D-vitamin bidrar till normalt kalciumupptag i tarmen. D-vitamin spelar också en roll i celldelningsprocessen och bidrar till ett normalt fungerande immunförsvar. En studie publicerad i American Journal of Clinical Nutrition drog slutsatsen att intag av grönt te ökade fettförbränningen med 4 % under 24 timmar! Craveless innehåller vitamin B1 (tiamin), B2 (riboflavin), B3 (niacin), B5 (pantotensyra), B6 (pyridoxin), B7 (biotin), B9 (folsyra) och B12. Vitamin B3 bidrar till normal energiomsättning och minskar trötthet och utmattning. Vitamin B6 bidrar till normal protein- och glykogenmetabolism, till normal bildning av röda blodkroppar, till normal energimetabolism och till reglering av hormonaktivitet. Vitamin B7 bidrar till normal makronäringsämnesomsättning och normal energiomsättning. Vitamin B9 spelar en roll i celldelningsprocessen och bidrar till normal aminosyrasyntes och normal homocysteinmetabolism. Vitamin B12 bidrar till nervsystemets normala funktion och till normal bildande av röda blodkroppar. Craveless on sinulle, joka haluat pitää makeanhimon kurissa yes tehdä parempien valintojen tekemisestä helpompaa. Craveless voi auttaa pitämään verensokerin tasaisena yes siten vähentää makeanhimoa. Samalla Craveless sisältää vihreästä teestä peräisin olevaa uutetta, joka lisää energiankulutusta, sekä B6- yes B12-vitaminja, jotka lisäävät energiaa. Aloita hyvät tavat jo tänään ja tilaa kokeilupakkaus 30 dagar. Kokeile hintaan 7,9 eur eli postikulujen hinnalla. ## [Klicka här för att beställa från Craveless officiella webbplats](https://craveless.com.se)
legenduck/test12	legenduck	2025-05-02T09:42:40Z	0	0	transformers	[ "transformers", "safetensors", "text-generation-inference", "unsloth", "llama", "trl", "en", "base_model:unsloth/Llama-3.2-1B-unsloth-bnb-4bit", "base_model:finetune:unsloth/Llama-3.2-1B-unsloth-bnb-4bit", "license:apache-2.0", "endpoints_compatible", "region:us" ]	null	2025-05-02T09:40:44Z	--- base_model: unsloth/Llama-3.2-1B-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - llama - trl license: apache-2.0 language: - en --- # Uploaded model - Developed by: legenduck - License: apache-2.0 - Finetuned from model : unsloth/Llama-3.2-1B-unsloth-bnb-4bit This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
legenduck/test13	legenduck	2025-05-02T09:41:39Z	0	0	transformers	[ "transformers", "safetensors", "text-generation-inference", "unsloth", "llama", "trl", "en", "base_model:unsloth/Llama-3.2-1B-unsloth-bnb-4bit", "base_model:finetune:unsloth/Llama-3.2-1B-unsloth-bnb-4bit", "license:apache-2.0", "endpoints_compatible", "region:us" ]	null	2025-05-02T09:39:43Z	--- base_model: unsloth/Llama-3.2-1B-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - llama - trl license: apache-2.0 language: - en --- # Uploaded model - Developed by: legenduck - License: apache-2.0 - Finetuned from model : unsloth/Llama-3.2-1B-unsloth-bnb-4bit This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
sanchit42/llama3.1-tokenizer-augmented-all_llama_3b	sanchit42	2025-05-02T09:38:12Z	0	0	transformers	[ "transformers", "arxiv:1910.09700", "endpoints_compatible", "region:us" ]	null	2025-05-02T09:38:07Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
maksf8486/16c86859-ce88-4d8e-a0c9-8bbdaea7c78f	maksf8486	2025-05-02T09:36:24Z	0	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:sethuiyer/Medichat-Llama3-8B", "base_model:adapter:sethuiyer/Medichat-Llama3-8B", "license:other", "8-bit", "bitsandbytes", "region:us" ]	null	2025-05-02T09:08:10Z	--- library_name: peft license: other base_model: sethuiyer/Medichat-Llama3-8B tags: - axolotl - generated_from_trainer model-index: - name: 16c86859-ce88-4d8e-a0c9-8bbdaea7c78f results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml absolute_data_files: false adapter: lora base_model: sethuiyer/Medichat-Llama3-8B bf16: true chat_template: llama3 dataset_prepared_path: null datasets: - data_files: - ebaa36ac6b1bdb65_train_data.json ds_type: json format: custom path: /workspace/input_data/ebaa36ac6b1bdb65_train_data.json type: field_input: reasoning (reasoning_content) field_instruction: question field_output: response (content) format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null dpo: beta: 0.1 enabled: true group_by_length: false rank_loss: false reference_model: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 1 gradient_checkpointing: true gradient_clipping: 0.5 group_by_length: false hub_model_id: maksf8486/16c86859-ce88-4d8e-a0c9-8bbdaea7c78f hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-06 load_in_4bit: false load_in_8bit: true local_rank: null logging_steps: 1 lora_alpha: 64 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 32 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 8 mixed_precision: bf16 mlflow_experiment_name: /tmp/ebaa36ac6b1bdb65_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 33f6b38d-f8bd-4301-b3c9-673be809902f wandb_project: s56-2 wandb_run: your_name wandb_runid: 33f6b38d-f8bd-4301-b3c9-673be809902f warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 16c86859-ce88-4d8e-a0c9-8bbdaea7c78f This model is a fine-tuned version of [sethuiyer/Medichat-Llama3-8B](https://huggingface.co/sethuiyer/Medichat-Llama3-8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.8439 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-06 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.9092 \| 0.0802 \| 200 \| 0.8439 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
Romain-XV/bb8374c7-21c2-45f7-9717-b7b722841c1d	Romain-XV	2025-05-02T09:36:15Z	0	0	transformers	[ "transformers", "pytorch", "tensorboard", "safetensors", "llama", "text-generation", "generated_from_trainer", "axolotl", "dpo", "trl", "conversational", "arxiv:2305.18290", "base_model:The-matt/llama2_ko-7b_distinctive-snowflake-182_1060", "base_model:finetune:The-matt/llama2_ko-7b_distinctive-snowflake-182_1060", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]	text-generation	2025-05-02T08:41:13Z	--- base_model: The-matt/llama2_ko-7b_distinctive-snowflake-182_1060 library_name: transformers model_name: bb8374c7-21c2-45f7-9717-b7b722841c1d tags: - generated_from_trainer - axolotl - dpo - trl licence: license --- # Model Card for bb8374c7-21c2-45f7-9717-b7b722841c1d This model is a fine-tuned version of [The-matt/llama2_ko-7b_distinctive-snowflake-182_1060](https://huggingface.co/The-matt/llama2_ko-7b_distinctive-snowflake-182_1060). It has been trained using [TRL](https://github.com/huggingface/trl). ## Quick start ```python from transformers import pipeline question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?" generator = pipeline("text-generation", model="Romain-XV/bb8374c7-21c2-45f7-9717-b7b722841c1d", device="cuda") output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0] print(output["generated_text"]) ``` ## Training procedure [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/romain_fnc-xventures/Gradients-On-Demand/runs/2c5h2sts) This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290). ### Framework versions - TRL: 0.12.0.dev0 - Transformers: 4.46.0 - Pytorch: 2.5.0+cu124 - Datasets: 3.0.1 - Tokenizers: 0.20.1 ## Citations Cite DPO as: ```bibtex @inproceedings{rafailov2023direct, title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}}, author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn}, year = 2023, booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023}, url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html}, editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine}, } ``` Cite TRL as: ```bibtex @misc{vonwerra2022trl, title = {{TRL: Transformer Reinforcement Learning}}, author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec}, year = 2020, journal = {GitHub repository}, publisher = {GitHub}, howpublished = {\url{https://github.com/huggingface/trl}} } ```
unsloth/GLM-4-32B-0414-GGUF	unsloth	2025-05-02T09:35:27Z	11,192	16	transformers	[ "transformers", "gguf", "glm4", "text-generation", "unsloth", "zh", "en", "base_model:THUDM/GLM-4-32B-0414", "base_model:quantized:THUDM/GLM-4-32B-0414", "license:mit", "autotrain_compatible", "endpoints_compatible", "region:us", "imatrix", "conversational" ]	text-generation	2025-04-25T09:54:10Z	--- tags: - unsloth base_model: - THUDM/GLM-4-32B-0414 license: mit language: - zh - en pipeline_tag: text-generation library_name: transformers --- <p style="margin-top: 0;"> <strong><a href="https://docs.unsloth.ai/basics/unsloth-dynamic-v2.0-gguf">Unsloth Dynamic 2.0</a> achieves superior accuracy & outperforms other leading quants.</strong> </p> # GLM-4-32B-0414 ## Introduction The GLM family welcomes new members, the GLM-4-32B-0414 series models, featuring 32 billion parameters. Its performance is comparable to OpenAI’s GPT series and DeepSeek’s V3/R1 series. It also supports very user-friendly local deployment features. GLM-4-32B-Base-0414 was pre-trained on 15T of high-quality data, including substantial reasoning-type synthetic data. This lays the foundation for subsequent reinforcement learning extensions. In the post-training stage, we employed human preference alignment for dialogue scenarios. Additionally, using techniques like rejection sampling and reinforcement learning, we enhanced the model’s performance in instruction following, engineering code, and function calling, thus strengthening the atomic capabilities required for agent tasks. GLM-4-32B-0414 achieves good results in engineering code, Artifact generation, function calling, search-based Q&A, and report generation. In particular, on several benchmarks, such as code generation or specific Q&A tasks, GLM-4-32B-Base-0414 achieves comparable performance with those larger models like GPT-4o and DeepSeek-V3-0324 (671B). GLM-Z1-32B-0414 is a reasoning model with deep thinking capabilities. This was developed based on GLM-4-32B-0414 through cold start, extended reinforcement learning, and further training on tasks including mathematics, code, and logic. Compared to the base model, GLM-Z1-32B-0414 significantly improves mathematical abilities and the capability to solve complex tasks. During training, we also introduced general reinforcement learning based on pairwise ranking feedback, which enhances the model's general capabilities. GLM-Z1-Rumination-32B-0414 is a deep reasoning model with rumination capabilities (against OpenAI's Deep Research). Unlike typical deep thinking models, the rumination model is capable of deeper and longer thinking to solve more open-ended and complex problems (e.g., writing a comparative analysis of AI development in two cities and their future development plans). Z1-Rumination is trained through scaling end-to-end reinforcement learning with responses graded by the ground truth answers or rubrics and can make use of search tools during its deep thinking process to handle complex tasks. The model shows significant improvements in research-style writing and complex tasks. Finally, GLM-Z1-9B-0414 is a surprise. We employed all the aforementioned techniques to train a small model (9B). GLM-Z1-9B-0414 exhibits excellent capabilities in mathematical reasoning and general tasks. Its overall performance is top-ranked among all open-source models of the same size. Especially in resource-constrained scenarios, this model achieves an excellent balance between efficiency and effectiveness, providing a powerful option for users seeking lightweight deployment. ## Showcase ### Animation Generation <table> <tr> <td style="text-align: center; font-size: 16px; font-weight: bold; padding: 10px; width: 420px;"> GLM-Z1-32B-0414 </td> <td style="text-align: center; font-size: 16px; font-weight: bold; padding: 10px; width: 420px;"> GLM-4-32B-0414 </td> </tr> <tr> <td style="vertical-align: top; padding: 10px; width: 420px;"> <video src="https://github.com/user-attachments/assets/849ff9fd-b54d-4c74-9ee5-3412e1a09e32" style="width: 400px; height: 300px; object-fit: contain;" autoplay loop muted playsinline></video> <div style="margin-top: 10px; font-size: 14px; color: #333; width: 400px;"> write a Python program that shows a ball bouncing inside a spinning hexagon. The ball should be affected by gravity and friction, and it must bounce off the rotating walls realistically </div> </td> <td style="vertical-align: top; padding: 10px; width: 420px;"> <video src="https://github.com/user-attachments/assets/8dccdb9d-cc44-4732-b438-74a4e3cb9dfb" style="width: 400px; height: 300px; object-fit: contain;" autoplay loop muted playsinline></video> <div style="margin-top: 10px; font-size: 14px; color: #333; width: 400px;"> Use HTML to simulate the scenario of a small ball released from the center of a rotating hexagon. Consider the collision between the ball and the hexagon's edges, the gravity acting on the ball, and assume all collisions are perfectly elastic. (Prompt translated from Chinese) </div> </td> </tr> </table> ### Web Design <table> <tr> <td style="text-align: center; font-size: 16px; font-weight: bold; padding: 10px; width: 420px;"> GLM-4-32B-0414 </td> <td style="text-align: center; font-size: 16px; font-weight: bold; padding: 10px; width: 420px;"> GLM-4-32B-0414 </td> </tr> <tr> <td style="vertical-align: top; padding: 10px; width: 420px;"> <img src="https://github.com/user-attachments/assets/bd9c1fc1-c784-4e8f-9c76-5f7389a715f1"/> <div style="margin-top: 10px; font-size: 14px; color: #333; width: 400px;"> Design a drawing board that supports custom function plotting, allowing adding and deleting custom functions, and assigning colors to functions. (Prompt translated from Chinese) </div> </td> <td style="vertical-align: top; padding: 10px; width: 420px;"> <img src="https://github.com/user-attachments/assets/7ad12d52-9229-4278-8d1b-ffbf43e99070"/> <div style="margin-top: 10px; font-size: 14px; color: #333; width: 400px;"> Design a UI for a mobile machine learning platform, which should include interfaces for training tasks, storage management, and personal statistics. The personal statistics interface should use charts to display the user's resource usage over a period. Use Tailwind CSS to style the page, and display these 3 mobile interfaces tiled on a single HTML page. (Prompt translated from Chinese) </div> </td> </tr> </table> ### SVG Generation <table> <tr> <td style="text-align: center; font-size: 16px; font-weight: bold; padding: 10px; width: 420px;"> GLM-4-32B-0414 </td> <td style="text-align: center; font-size: 16px; font-weight: bold; padding: 10px; width: 420px;"> GLM-4-32B-0414 </td> </tr> <tr> <td style="vertical-align: top; padding: 10px; width: 420px;"> <img src="https://github.com/user-attachments/assets/9407e4c1-1876-4ab5-838c-839836fb418a"/> <div style="margin-top: 10px; font-size: 14px; color: #333; width: 400px;"> Create a misty Jiangnan scene using SVG. (Prompt translated from Chinese) </div> </td> <td style="vertical-align: top; padding: 10px; width: 420px;"> <img src="https://github.com/user-attachments/assets/bcce8c5a-cedf-45c8-b666-ddb023d5b49c"/> <div style="margin-top: 10px; font-size: 14px; color: #333; width: 400px;"> Use SVG to illustrate the training process of an LLM. (Prompt translated from Chinese) </div> </td> </tr> </table> ### Search-Based Writing For search-based writing tasks, we use the following system prompt to have the model respond based on search results: ``` 请根据所给搜索返回结果对用户问题进行作答。 ## 注意 1. 充分利用和整理收集到的信息，而不是简单的复制粘贴，生成符合用户要求且有深度的专业答案。 2. 所提供信息充分的情况下，你的回答需尽可能延长，从用户意图角度出发，提供具有足够信息量和多角度的回复。 3. 另外，并非所有的搜索结果都与用户问题密切相关，请仔细的甄别、筛选和利用。 4. 客观类问答的答案通常非常简短，你可以适当补充一到两句相关信息，以丰富内容。 5. 请确保你的回复格式美观、可读性强。对于多实体对比或列举，善用列表格式来帮助用户更好的理解信息。 6. 除非用户要求，否则你回答的语言请于用户提问语言保持一致。 7. 在适当情况下在句子末尾使用例如:【0†source】的格式引用搜索结果。 ``` When using, you can obtain search results through methods such as `RAG` or `WebSearch`, and wrap them in `observation`, for example: ```json [ { "role": "user", "content": "Explore the common characteristics of children's literature, with a focus on its narrative techniques and thematic tendencies. This includes narrative techniques: common approaches in children's literature such as first-person, third-person, omniscient narrator, and interactive narration, and their influence on young readers. It also includes thematic tendencies: recurring themes in children's literature such as growth, adventure, friendship, and family, with an analysis of how these themes impact children's cognitive and emotional development. Additionally, other universal features such as the use of personification, repetitive language, symbolism and metaphor, and educational value should be considered. Please provide a detailed analytical report based on academic research, classic examples of children's literature, and expert opinions." }, { "role": "observation", "content": "【{id}†{title}†{url}】\n{content}" }, ... ] ``` For the above prompt, we use an internal or external search model to obtain the search results. Using the format shown above, we can generate the following analysis report: <div style="height: 400px; width: 100%; overflow: auto; border: 5px solid #ddd; padding: 20px;"> # Analysis Report on Common Characteristics of Children's Literature Children's literature, as a literary genre specifically created for children, possesses unique artistic features and expressive techniques. This report will comprehensively analyze the common characteristics of children's literature from three dimensions: narrative methods, thematic tendencies, and other universal features, integrating academic research, classic examples of children's literature, and expert viewpoints. ## Narrative Methods and Their Impact on Child Readers The narrative methods of children's literature exhibit diverse characteristics, including first-person, third-person, narration, and interactive storytelling. These different narrative perspectives and methods profoundly influence children's reading experiences and cognitive development. First-person narration is common in children's literature, unfolding the story from the perspective of a child protagonist. This narrative style bridges the gap between the reader and the story, making it easier for child readers to identify with the characters and develop emotional resonance. For example, many coming-of-age novels use first-person narration, allowing the protagonist to directly recount their experiences and feelings, making readers feel as though they are experiencing the joys and sorrows of growth alongside the protagonist. This narrative perspective lends the work a strong sense of authenticity and intimacy, helping to cultivate children's empathy【1†source】. Third-person narration offers a broader perspective, allowing the author to flexibly switch between different characters' viewpoints and present richer layers of the story. In children's literature, third-person omniscient narration enables the author to control the narrative pace, revealing or concealing information as needed to guide children's attention. At the same time, third-person narration facilitates direct dialogue between the author and the reader, conveying values or explaining complex concepts through narration. This narrative method positively influences children's macro-thinking and comprehensive understanding【1†source】. Narration (authorial intrusion) is a unique narrative technique in children's literature, where the author directly appears as the "storyteller," explaining the background, commenting on characters, or posing questions to the reader. This technique is particularly common in classic fairy tales, such as the opening lines of Andersen's Fairy Tales: "Once, there was a child..." Narration helps children understand the story's context, fills cognitive gaps, and conveys the author's educational intent. Research shows that appropriate authorial intrusion aids children in grasping the story's structure and improving reading comprehension【5†source】. Interactive storytelling is a new trend in contemporary children's literature, especially prominent in the digital media era. Interactive storytelling breaks the traditional unidirectional author-reader relationship, encouraging child readers to participate in the story's creation, such as by choosing plot directions, character dialogues, or endings. This participatory reading enhances children's sense of agency and fosters decision-making skills and creative thinking. For example, some children's reading apps incorporate interactive elements, allowing children to influence the story's development through clicks, drag-and-drop actions, and other operations, thereby gaining a stronger sense of immersion and achievement【6†source】. Interactive storytelling transforms children from passive information recipients into active meaning-makers, uniquely contributing to the development of their subjectivity. Table: Common Narrative Methods in Children's Literature and Their Effects \| Narrative Method \| Characteristics \| Impact on Child Readers \| Classic Examples \| \|----------------------\|--------------------\|----------------------------\|---------------------\| \| First-Person \| Told from the child protagonist's perspective \| Enhances immersion, fosters empathy \| Charlotte's Web, The Straw House \| \| Third-Person \| Omniscient or limited perspective \| Expands horizons, develops comprehensive understanding \| Harry Potter series \| \| Narration \| Direct authorial intrusion into the narrative \| Aids comprehension, conveys values \| Andersen's Fairy Tales \| \| Interactive \| Encourages reader participation in creation \| Cultivates agency and creative thinking \| Children's interactive reading apps \| Notably, the narrative methods of children's literature are often closely intertwined with the childhood perspective. The childhood perspective does not necessarily mean the narrator must be a child but refers to the work's ability to describe the world to the greatest extent from a child's heart, expressing their inner psychology and external circumstances【2†source】. Through the childhood perspective, readers can embark on a spiritual journey with a child's mindset, a narrative strategy that creates a strong sense of realism, allowing child readers to achieve emotional identification and cognitive resonance during the reading process【1†source】. The use of the childhood perspective gives the work's language a perceptual and naive quality, often with a prose-like and spatial structure, artistic features that align with children's cognitive characteristics and aid their acceptance and understanding【2†source】. ## Thematic Tendencies and Their Impact on Children's Cognitive and Emotional Development The thematic choices in children's literature exhibit distinct tendencies, with common themes including growth, adventure, friendship, and family. These themes not only form the core content of children's literature but also subtly influence children's cognitive development and emotional shaping. The theme of growth is one of the central motifs in children's literature. Growth narratives are regarded as the artistic lifeblood of children's literature, focusing on depicting the pivotal moments of rapid psychological development in children, particularly the awakening and establishment of self-awareness【3†source】. Growth literature typically includes three elements: an artistic portrayal of the self-awareness construction process in growing adolescents, a developmental story with logical propulsion, and the presentation of the protagonist's spiritual trials and quest for direction【3†source】. By reading growth-themed works, child readers can indirectly experience the confusion and breakthroughs of growing up and understand the formation of self-identity. Classics such as Astrid Lindgren's Pippi Longstocking and Cao Wenxuan's The Straw House vividly depict children's psychological growth trajectories in specific environments. Research indicates that growth-themed literary works help children build a positive self-concept and develop the courage and resilience to face challenges, positively contributing to their psychological development【9†source】. The theme of adventure holds an important place in children's literature, satisfying children's curiosity about exploring the unknown. Adventure stories often feature unusual settings and unknown challenges, with the protagonist growing through overcoming difficulties. Classics like Robinson Crusoe and The Adventures of Tom Sawyer attract child readers with thrilling plots while conveying the importance of qualities such as courage, wisdom, and perseverance. The impact of adventure themes on children's cognitive development mainly lies in expanding their imaginative space and fostering problem-solving skills. In adventure stories, children must analyze situations, make plans, and respond to unexpected events alongside the protagonist, a process that exercises their logical thinking and adaptability【14†source】. At the same time, the unfamiliar environments and novel experiences in adventure stories stimulate children's curiosity and desire to learn, laying the foundation for cultivating an exploratory spirit. As experts point out, excellent children's literature should be grounded in reality, rich in depth, and generate significant inspiration and感染力, guiding children to comprehensively understand the world【14†source】. The theme of friendship is equally prevalent in children's literature, reflecting children's emphasis on peer relationships. Friendship and love are regarded as humanity's most precious qualities, often depicted in children's literature as beacons in the night, guiding children toward the future【9†source】. Friendship stories typically revolve around interactions between children, portraying positive behaviors such as sharing, cooperation, and understanding. Examples include the genuine friendships among the children at Tomoe Gakuen in Totto-Chan: The Little Girl at the Window and the promise and mutual aid between Wilbur and Charlotte in Charlotte's Web. These stories help child readers recognize the value of friendship and learn how to build and maintain interpersonal relationships. Research shows that children need peer support during their growth, as friends provide crucial emotional anchors, offering the greatest emotional support and comfort in unfamiliar environments【16†source】. By reading friendship-themed works, children can learn social skills, develop empathy, and cultivate a spirit of cooperation, qualities essential for their social development【17†source】. The theme of family is an indispensable subject in children's literature, depicting the emotional bonds and interaction patterns among family members. As the primary setting for children's earliest socialization, the family atmosphere and parenting styles profoundly impact children's mental health【10†source】. Family stories in children's literature often focus on parent-child relationships, sibling bonds, and other dynamics, such as Alice's relationship with her sister in Alice's Adventures in Wonderland and the Little Prince's interactions with the rose in The Little Prince. These stories help children understand the responsibilities and expectations of family roles and learn to handle conflicts within the family. Research indicates that a positive family atmosphere and parental support promote the development of children's positive psychological traits, while adverse family environments and parenting behaviors negatively affect their mental health【10†source】【11†source】. By reading family-themed works, children can gain emotional support, learn skills for managing family relationships, and establish healthy family values. Table: Common Themes in Children's Literature and Their Impact on Child Development \| Theme Type \| Content Representation \| Impact on Cognitive Development \| Impact on Emotional Development \| Classic Examples \| \|---------------\|---------------------------\|-------------------------------------\|-------------------------------------\|---------------------\| \| Growth \| Awakening of self-awareness, psychological trials and breakthroughs \| Establishes self-concept, fosters problem-solving skills \| Shapes positive self-identity, enhances psychological resilience \| The Straw House, Pippi Longstocking \| \| Adventure \| Exploring the unknown, overcoming challenges \| Expands imaginative space, exercises logical thinking \| Cultivates courage and perseverance \| Robinson Crusoe, The Adventures of Tom Sawyer \| \| Friendship \| Peer interactions, mutual aid and cooperation \| Learns social skills, understands interpersonal dynamics \| Develops empathy, builds a sense of belonging \| Charlotte's Web, Totto-Chan: The Little Girl at the Window \| \| Family \| Parent-child relationships, sibling bonds \| Understands social roles, learns communication skills \| Gains emotional support, establishes secure attachments \| Alice's Adventures in Wonderland, The Little Prince \| Regarding thematic choices, children's literature researcher Zhu Ziqiang proposed the famous "Three Major Motifs" theory, categorizing children's literary works into "the motif of love," "the motif of the mischievous child," and "the motif of nature"【8†source】. The motif of love focuses on emotional connections between children and adults or peers; the motif of the mischievous child portrays children's free-spirited nature; and the motif of nature emphasizes the harmonious relationship between children and the natural environment. These three motifs reflect the richness of the children's world from different angles, providing diverse emotional experiences and cognitive frameworks for children. Notably, these themes do not exist in isolation; outstanding works often organically integrate multiple themes. For example, the Harry Potter series incorporates growth, friendship, adventure, and family elements, presenting child readers with a multidimensional spiritual world. ## Other Universal Features and Their Artistic Expression In addition to narrative methods and thematic tendencies, children's literature exhibits a series of universal artistic features, including anthropomorphism, repetitive language, symbolism and metaphor, and educational significance. These features collectively constitute the unique aesthetic style of children's literature, subtly influencing children's cognitive development and aesthetic cultivation. Anthropomorphism is one of the most distinctive artistic features of children's literature. In children's literary works, animals, plants, and even inanimate objects are often endowed with human thoughts, emotions, and behaviors, greatly enhancing the story's fun and imagination. Research shows that anthropomorphism is a frequently used technique by children's literature creators to attribute human characteristics to animals, enabling them to possess perception and communication abilities【19†source】. Through anthropomorphism, children can more easily understand abstract concepts and moral principles, as anthropomorphic characters translate complex ideas into familiar emotional and behavioral patterns. For example, in scientific fairy tales, anthropomorphic characters can help explain scientific principles, making abstract concepts tangible【18†source】. Anthropomorphism not only enriches the narrative techniques of children's literature but also provides children with a unique perspective for understanding the relationship between humans and nature. It is worth noting that excessive anthropomorphism may affect children's accurate understanding of the animal world, so modern children's literature pays more attention to balancing the natural attributes of characters with human characteristics when employing anthropomorphic techniques【19†source】. Repetitive language is extremely common in children's literature, a linguistic feature rooted in oral traditions originally intended to aid memory and dissemination【20†source】. In children's literature, the repetitive use of words, phrases, or sentences serves multiple functions: constructing the story's framework, emphasizing key information, creating rhythm and musicality, and training children's vocabulary skills. For example, in The Very Hungry Caterpillar, the author repeatedly uses phrases like "On Monday, he ate one apple. On Tuesday, he ate two pears..." This not only builds the story's structure but also helps children learn numbers and days of the week. Repetitive structures also aid children in developing an awareness of language patterns during the early stages of language acquisition, fostering a sense of language and memory skills【21†source】. Research indicates that repetitive language in children's literature promotes children's language acquisition, helping them master vocabulary and syntactic rules. At the same time, this linguistic feature enhances the story's participatory nature, as children can often join in reciting the repetitive parts, gaining a sense of achievement. Symbolism and metaphor are common expressive techniques in children's literature, conveying abstract meanings through concrete imagery. Symbolism uses specific objects to represent abstract concepts or emotions, while metaphor connects two different things through comparison, creating new meanings. In children's literature, symbolism and metaphor are usually presented in a simple and clear manner, avoiding overly complex interpretations. For example, the character configurations and metaphorical connotations in The Wizard of Oz are thought-provoking, as these characters not only breathe life into the story but also convey profound life philosophies through their symbolic meanings【24†source】. Symbolism and metaphor in children's literature are often related to themes such as growth, friendship, and courage, helping children understand abstract concepts through concrete and figurative expressions. Research shows that appropriate metaphors can promote children's cognitive development, stimulating their imagination and creativity【23†source】. As children grow older, their ability to understand symbolism and metaphor gradually improves, providing children's literature with multi-layered meaning spaces. Educational significance is an indispensable component of children's literature, which inherently carries the gene of children's education【22†source】. Excellent children's literary works simultaneously possess entertainment and educational functions, not only helping children understand the objective world, enrich their inner emotions, and acquire life wisdom but also cultivating their perception, aesthetic sensibility, thinking skills, and creativity【15†source】. Educational significance in children's literature is often not directly presented through preaching but naturally revealed through the storyline and characters' fates. For example, many classic fairy tales convey the importance of qualities such as bravery and honesty through the protagonist's adventurous experiences, while popular science books introduces scientific knowledge through interesting plots and characters. Experts point out that children's literature writers should shoulder the importantence of education, incorporating care for children's mental growth into their works【22†source】. It is worth noting that the educational significance of children's literature should respect children's receptive abilities, avoiding excessive preaching or moral indoctrination, and instead naturally influencing children's values and behaviors through artistic appeal. Storytelling is the most basic and essential feature of children's literature. Children's perceptual, imagery-driven, and novelty-seeking cognitive characteristics and receptive psychology further determine that "storytelling" is an indispensable ontological feature of children's literature【25†source】. Engaging plots are the most crucial aspect of children's literary works because, compared to adults, children's understanding of things relies mainly on intuition, and plots play a key role in guiding children's comprehension of stories【26†source】. The storytelling quality of children's literature is reflected in multiple aspects: clear cause-and-effect relationships, Compact narrative rhythm and satisfying endings. These elements work together to immerse children in the story world, providing emotional satisfaction and cognitive inspiration. As researchers have noted, plots must be performed by specific characters in specific situations to convey individual experiences in unique space-time environments【7†source】. In children's literature, storytelling is not merely an artistic technique but a bridge connecting children to the world. Through stories, children can safely experience various life scenarios and learn methods for challenges. In terms of language features, children's literature typically adopts a concise, clear, and vivid language style, avoiding complex sentence structures and abstract vocabulary. This linguistic characteristic aligns with children's cognitive development levels, facilitating their understanding and acceptance. At the same time, the language of children's literature is often rich in rhythm and musicality, enhancing readability and memorability through techniques such as rhyming and repetition. For example, Michael Rosen's children's literary works extensively employ repetitive structures and rhymes, a language usage that helps children develop an awareness of language patterns during the early stages of language acquisition【21†source】. The language of children's literature also often includes rich sensory descriptions and emotional expressions, stimulating children's imagination through concrete and tangible imagery. Scholar Jay Davis's research shows that the interactive use of language in children's literature can influence children's language habits and promote their language development【21†source】. In summary, these universal features of children's literature collectively constitute its unique artistic charm and educational value. Anthropomorphism and symbolism expand children's imaginative spaces, repetitive language and storytelling promote language acquisition and cognitive development, and the natural integration of educational significance achieves the artistic effect of "teaching through entertainment." These features do not exist in isolation but are interwoven and organically unified, collectively serving the comprehensive development of child readers. ## Conclusion Through a systematic analysis of the narrative methods, thematic tendencies, and other universal features of children's literature, we can draw the following conclusions: As a special literary genre, the creation and reception of children's literature follow unique rules. In terms of narrative methods, children's literature flexibly employs various techniques such as first-person, third-person, narration, and interactive storytelling to adapt to children's cognitive characteristics and receptive psychology. Among these, the use of the childhood perspective is particularly important, as it enhances the work's sense of realism and intimacy, enabling child readers to develop emotional resonance【1†source】【2†source】. In terms of thematic choices, growth, adventure, friendship, and family constitute the main content of children's literature. These themes not only satisfy children's curiosity and desire to explore but also subtly influence their cognitive development and emotional shaping【3†source】【9†source】. Other universal features such as anthropomorphism, repetitive language, symbolism, and educational significance collectively form the unique artistic style and educational value of children's literature【18†source】【20†source】【24†source】. These characteristics of children's literature do not exist in isolation but are interconnected and organically unified. For example, adventure themes are often combined with third-person omniscient narration to attract child readers through compact plots and vivid descriptions; friendship themes frequently employ first-person narration to enhance emotional resonance; and anthropomorphism is commonly found in nature-themed works, helping children understand the relationship between humans and nature. These features collectively serve the comprehensive development of child readers, meeting their entertainment needs while promoting their cognitive growth and emotional maturity. From an academic research perspective, children's literature studies should emphasize the application of narrative theory, as narrative theory focuses more on the "how" of storytelling—narrative form—which aligns closely with the research focus of children's literature【0†source】. At the same time, cognitive research methods provide new perspectives for children's literature studies. By combining cognitive science with literary theory, we can gain a deeper understanding of how children's literature influences children's thinking and cognitive development【4†source】. Future research should continue to explore the application of these theoretical methods in children's literature studies while paying attention to the intersection and integration of children's literature with emerging fields such as digital media and interdisciplinary education. From a creative practice perspective, children's literature writers should fully grasp children's cognitive characteristics and emotional needs, incorporating growth Care and educational wisdom into their work As experts have pointed out, excellent children's literary works should be grounded in reality, rich in depth, and generate significant infection and infectivity, guiding children to comprehensively understand the world and correctly recognize themselves and society【14†source】. At the same time, children's literature Creativity should keep pace with the times, addressing new problems and challenges faced by contemporary children, such as media literacy in the digital age and identity formation in multicultural contexts, to provide targeted spiritual nourishment for children. From an educational application perspective, children's literature should fully leverage its unique role in children's mental growth. Through carefully designed reading activities, teachers and parents can help children deeply understand the themes and meanings in works, guiding them to connect reading experiences with real life. Research shows that children's literature plays an increasingly important role in language education, the construction of a reading society, and children's mental growth【22†source】. Therefore, children's literature should be incorporated as an important component of school and family education, promoting children's cognitive development and emotional maturity through activities such as reading sharing, role-playing, and creative writing. In summary, as a unique art form and educational medium, the common characteristics of children's literature constitute an organic whole, collectively serving the comprehensive development of child readers. By deeply understanding these features and their mechanisms of influence, we can better create, research, and apply children's literature, providing high-quality spiritual nourishment for children's healthy growth. Future children's literature research should continue to deepen theoretical exploration, expand research methods, and strengthen interdisciplinary collaboration to address the ever-changing needs of children and the challenges of the times, promoting the continuous development of children's literature. </div> ### Function Call GLM-4-32B-0414 supports calling external tools in JSON format. This can be done via HuggingFace Transformers, vLLM, or sgLang. The message format for tool calling is as follows: ```json= { "role": "asssitant", "metadata": function_name, "content": json.dumps(call_arguments, ensure_ascii=False) } ``` The message format for tool execution results is as follows: ```json= { "role": "observation", "content": json.dumps(tool_response, ensure_ascii=False) if not isinstance(tool_response, str) else tool_response } ``` The following example demonstrates the process of GLM-4-32B-0414 calling a tool and generating a final response using HuggingFace Transformers. ```python import json import re import ast from transformers import AutoModelForCausalLM, AutoTokenizer MODEL_PATH = "THUDM/GLM-4-32B-0414" tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH) model = AutoModelForCausalLM.from_pretrained(MODEL_PATH, device_map="auto") def is_function_call(single_message): """Determine whether the current system message is a function call.""" pattern = re.compile(r'([^\n`]?)\n({.?})(?=\w\n\|$)', re.DOTALL) matches = pattern.findall(single_message) if not matches: return False func_name, args_str = matches[0] func_name = func_name.strip() try: parsed_args = json.loads(args_str) except json.JSONDecodeError: try: parsed_args = ast.literal_eval(args_str) except: return False return {"name": func_name, "arguments": parsed_args} def realtime_aqi(city): """Weather Query Tool""" if '北京' in city.lower(): return json.dumps({'city': '北京', 'aqi': '10', 'unit': 'celsius'}, ensure_ascii=False) elif '上海' in city.lower(): return json.dumps({'city': '上海', 'aqi': '72', 'unit': 'fahrenheit'}, ensure_ascii=False) else: return json.dumps({'city': city, 'aqi': 'unknown'}, ensure_ascii=False) def build_system_prompt(tools): """Construct system prompt based on the list of available tools.""" if tools is None: tools = [] value = "# 可用工具" contents = [] for tool in tools: content = f"\n\n## {tool['function']['name']}\n\n{json.dumps(tool['function'], ensure_ascii=False, indent=4)}" content += "\n在调用上述函数时，请使用 Json 格式表示调用的参数。" contents.append(content) value += "".join(contents) return value tools = [ { "type": "function", "function": { "name": "realtime_aqi", "description": "天气预报。获取实时空气质量。当前空气质量，PM2.5，PM10信息", "parameters": { "type": "object", "properties": { "city": { "description": "城市名" } }, "required": [ "city" ] } } } ] system_prompt = build_system_prompt(tools) message = [ {"role": "system", "content": system_prompt}, {"role": "user", "content": "北京和上海今天的天气情况"} ] print(f"User Message: {message[-1]['content']}") while True: inputs = tokenizer.apply_chat_template( message, return_tensors="pt", add_generation_prompt=True, return_dict=True, ).to(model.device) generate_kwargs = { "input_ids": inputs["input_ids"], "attention_mask": inputs["attention_mask"], "max_new_tokens": 1024, "do_sample": True, } out = model.generate(*generate_kwargs) generate_resp = tokenizer.decode(out[0][inputs["input_ids"].shape[1]:-1], skip_special_tokens=False) stop_sequence = tokenizer.decode(out[0][-1:], skip_speical_tokens=False) if stop_sequence == "<\|user\|>": print(f"Assistant Response: {generate_resp.strip()}") break function_calls = [] for m in generate_resp.split("<\|assistant\|>"): fc_decode = is_function_call(m.strip()) if fc_decode: message.append({"role": "assistant", "metadata": fc_decode['name'], "content": json.dumps(fc_decode['arguments'], ensure_ascii=False)}) print(f"Function Call: {fc_decode}") function_calls.append(fc_decode) else: message.append({"role": "assistant", "content": m}) print(f"Assistant Response: {m.strip()}") for fc in function_calls: function_response = realtime_aqi( city=fc["arguments"]["city"], ) print(f"Function Response: {function_response}") message.append({"role": "observation", "content": function_response}) ``` ## Evaluation Results <div style="text-align: center;"> <img src="https://raw.githubusercontent.com/THUDM/GLM-4/refs/heads/main/resources/Bench-32B.png" style="width: 80%;" /> </div> ### GLM-4-0414 Series \| 模型 \| IFEval \| BFCL-v3 (Overall) \| BFCL-v3 (MultiTurn) \| TAU-Bench (Retail) \| TAU-Bench (Airline) \| SimpleQA \| HotpotQA \| \| ---------------- \| ------ \| ----------------- \| ------------------- \| ------------------ \| ------------------- \| -------- \| -------- \| \| Qwen2.5-Max \| 85.6 \| 50.9 \| 30.5 \| 58.3 \| 22.0 \| 79.0 \| 52.8 \| \| GPT-4o-1120 \| 81.9 \| 69.6 \| 41.0 \| 62.8 \| 46.0 \| 82.8 \| 63.9 \| \| DeepSeek-V3-0324 \| 83.4 \| 66.2 \| 35.8 \| 60.7 \| 32.4 \| 82.6 \| 54.6 \| \| DeepSeek-R1 \| 84.3 \| 57.5 \| 12.4 \| 33.0 \| 37.3 \| 83.9 \| 63.1 \| \| GLM-4-32B-0414 \| 87.6 \| 69.6 \| 41.5 \| 68.7 \| 51.2 \| 88.1 \| 63.8 \| > For `SimpleQA` and `HotpotQA`, we sampled nearly 500 test cases from each test set, provided all models with basic `search` and `click` tools, ensured other settings remained consistent, and averaged the results over 3 runs. \| Model \| Framework \| [SWE-bench Verified](https://openai.com/index/introducing-swe-bench-verified/) \| [SWE-bench Verified mini](https://github.com/mariushobbhahn/SWEBench-verified-mini) \| \|---\|---\|---\|---\| \| GLM-4-32B-0414 \| Moatless<sup>[1]</sup> \| 33.8 \| 38.0 \| \| GLM-4-32B-0414 \| Agentless<sup>[2]</sup> \| 30.7 \| 34.0 \| \| GLM-4-32B-0414 \| OpenHands<sup>[3]</sup> \| 27.2 \| 28.0 \| [1] [Moatless v0.0.3](https://github.com/aorwall/moatless-tools) used the following parameters: `response_format="react", thoughts_in_action=False, max_interations=30`. No retries on failed trajectories; other settings are default. [2] [Agentless v1.5.0](https://github.com/OpenAutoCoder/Agentless) used [BGE](https://github.com/FlagOpen/FlagEmbedding/blob/master/README.md) as the embedding model and [FAISS](https://github.com/facebookresearch/faiss) for similarity search. To speed up patch verification while maintaining performance, the timeout for running a single instance was changed from the default 300s to 180s. [3] [OpenHands v0.29.1](https://github.com/All-Hands-AI/OpenHands/tree/main) did not use YaRN context extension but limited runs to a maximum of 60 iterations and summarized the history to prevent exceeding the 32K context limit. Summarization was configured as `llm_config="condenser", keep_first=1, max_size=32`. No retries on failed trajectories.
JK-art17/model-lora-thvblond-01	JK-art17	2025-05-02T09:31:18Z	0	0	null	[ "license:other", "region:us" ]	null	2025-05-02T08:48:36Z	--- license: other license_name: flux-1-dev-non-commercial-license license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md ---
nicolaadrah/physics_adapt_model_project	nicolaadrah	2025-05-02T09:30:14Z	0	0	transformers	[ "transformers", "text-generation-inference", "unsloth", "llama", "en", "license:apache-2.0", "endpoints_compatible", "region:us" ]	null	2025-05-02T09:30:11Z	--- base_model: unsloth/meta-llama-3.1-8b-unsloth-bnb-4bit tags: - text-generation-inference - transformers - unsloth - llama license: apache-2.0 language: - en --- # Uploaded finetuned model - Developed by: nicolaadrah - License: apache-2.0 - Finetuned from model : unsloth/meta-llama-3.1-8b-unsloth-bnb-4bit This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
arosidi/vit-base-oxford-iiit-pets	arosidi	2025-05-02T09:29:25Z	0	0	transformers	[ "transformers", "tensorboard", "safetensors", "vit", "image-classification", "generated_from_trainer", "base_model:google/vit-base-patch16-224", "base_model:finetune:google/vit-base-patch16-224", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]	image-classification	2025-05-02T09:19:29Z	--- library_name: transformers license: apache-2.0 base_model: google/vit-base-patch16-224 tags: - image-classification - generated_from_trainer metrics: - accuracy model-index: - name: vit-base-oxford-iiit-pets results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # vit-base-oxford-iiit-pets This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the pcuenq/oxford-pets dataset. It achieves the following results on the evaluation set: - Loss: 0.1865 - Accuracy: 0.9432 ## 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: 0.0003 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 5 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:--------:\| \| 0.38 \| 1.0 \| 370 \| 0.3080 \| 0.9242 \| \| 0.2037 \| 2.0 \| 740 \| 0.2364 \| 0.9350 \| \| 0.1495 \| 3.0 \| 1110 \| 0.2132 \| 0.9459 \| \| 0.1517 \| 4.0 \| 1480 \| 0.2060 \| 0.9432 \| \| 0.1501 \| 5.0 \| 1850 \| 0.2052 \| 0.9432 \| ### Framework versions - Transformers 4.51.3 - Pytorch 2.6.0+cu124 - Datasets 3.5.1 - Tokenizers 0.21.1
Candystarabs240/Boosted	Candystarabs240	2025-05-02T09:29:08Z	0	1	null	[ "license:apache-2.0", "region:us" ]	null	2025-05-02T09:29:08Z	--- license: apache-2.0 ---
TOMFORD79/Zata_29	TOMFORD79	2025-05-02T09:25:49Z	0	0	null	[ "any-to-any", "omega", "omegalabs", "bittensor", "agi", "license:mit", "region:us" ]	any-to-any	2025-05-02T08:54:23Z	--- license: mit tags: - any-to-any - omega - omegalabs - bittensor - agi --- This is an Any-to-Any model checkpoint for the OMEGA Labs x Bittensor Any-to-Any subnet. Check out the [git repo](https://github.com/omegalabsinc/omegalabs-anytoany-bittensor) and find OMEGA on X: [@omegalabsai](https://x.com/omegalabsai).
ngovanluu21999/kio998	ngovanluu21999	2025-05-02T09:21:45Z	0	0	null	[ "license:bigscience-openrail-m", "region:us" ]	null	2025-05-02T09:21:45Z	--- license: bigscience-openrail-m ---
infogeo/12687786-3dcd-46b6-b965-7da61afc37ea	infogeo	2025-05-02T09:18:48Z	0	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:sethuiyer/Medichat-Llama3-8B", "base_model:adapter:sethuiyer/Medichat-Llama3-8B", "license:other", "4-bit", "bitsandbytes", "region:us" ]	null	2025-05-02T09:10:55Z	--- library_name: peft license: other base_model: sethuiyer/Medichat-Llama3-8B tags: - axolotl - generated_from_trainer model-index: - name: 12687786-3dcd-46b6-b965-7da61afc37ea results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml absolute_data_files: false adapter: lora base_model: sethuiyer/Medichat-Llama3-8B bf16: true chat_template: llama3 dataset_prepared_path: /workspace/axolotl datasets: - data_files: - ebaa36ac6b1bdb65_train_data.json ds_type: json format: custom path: /workspace/input_data/ebaa36ac6b1bdb65_train_data.json type: field_input: reasoning (reasoning_content) field_instruction: question field_output: response (content) format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 1 gradient_checkpointing: true gradient_clipping: 0.55 group_by_length: false hub_model_id: infogeo/12687786-3dcd-46b6-b965-7da61afc37ea hub_repo: null hub_strategy: end hub_token: null learning_rate: 1.0e-06 load_in_4bit: true load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 64 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 32 lora_target_linear: true lr_scheduler: cosine max_steps: 150 micro_batch_size: 8 mixed_precision: bf16 mlflow_experiment_name: /tmp/ebaa36ac6b1bdb65_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 33f6b38d-f8bd-4301-b3c9-673be809902f wandb_project: s56-28 wandb_run: your_name wandb_runid: 33f6b38d-f8bd-4301-b3c9-673be809902f warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # 12687786-3dcd-46b6-b965-7da61afc37ea This model is a fine-tuned version of [sethuiyer/Medichat-Llama3-8B](https://huggingface.co/sethuiyer/Medichat-Llama3-8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.1754 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-06 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 150 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 1.038 \| 0.0601 \| 150 \| 1.1754 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
kimxxxx/mistral_r128_alpah256_batch16_gradient2_Ler6e-5_fulldataset_1250steps	kimxxxx	2025-05-02T09:16:51Z	0	0	transformers	[ "transformers", "safetensors", "arxiv:1910.09700", "endpoints_compatible", "region:us" ]	null	2025-05-02T09:16:38Z	--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - Developed by: [More Information Needed] - Funded by [optional]: [More Information Needed] - Shared by [optional]: [More Information Needed] - Model type: [More Information Needed] - Language(s) (NLP): [More Information Needed] - License: [More Information Needed] - Finetuned from model [optional]: [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - Repository: [More Information Needed] - Paper [optional]: [More Information Needed] - Demo [optional]: [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - Hardware Type: [More Information Needed] - Hours used: [More Information Needed] - Cloud Provider: [More Information Needed] - Compute Region: [More Information Needed] - Carbon Emitted: [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> BibTeX: [More Information Needed] APA: [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
kokovova/ef4b672b-57ca-4395-820e-47f5c7dc8be4	kokovova	2025-05-02T09:16:34Z	0	0	peft	[ "peft", "safetensors", "llama", "axolotl", "generated_from_trainer", "base_model:sethuiyer/Medichat-Llama3-8B", "base_model:adapter:sethuiyer/Medichat-Llama3-8B", "license:other", "4-bit", "bitsandbytes", "region:us" ]	null	2025-05-02T09:06:56Z	--- library_name: peft license: other base_model: sethuiyer/Medichat-Llama3-8B tags: - axolotl - generated_from_trainer model-index: - name: ef4b672b-57ca-4395-820e-47f5c7dc8be4 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) <details><summary>See axolotl config</summary> axolotl version: `0.4.1` ```yaml adapter: lora base_model: sethuiyer/Medichat-Llama3-8B bf16: true chat_template: llama3 dataset_prepared_path: /workspace/axolotl datasets: - data_files: - ebaa36ac6b1bdb65_train_data.json ds_type: json format: custom path: /workspace/input_data/ebaa36ac6b1bdb65_train_data.json type: field_input: reasoning (reasoning_content) field_instruction: question field_output: response (content) format: '{instruction} {input}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null early_stopping_patience: null eval_max_new_tokens: 128 eval_table_size: null evals_per_epoch: 1 flash_attention: true fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 1 gradient_checkpointing: true gradient_clipping: 0.5 group_by_length: false hub_model_id: kokovova/ef4b672b-57ca-4395-820e-47f5c7dc8be4 hub_repo: null hub_strategy: end hub_token: null learning_rate: 5.0e-06 load_in_4bit: true load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 64 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 32 lora_target_linear: true lr_scheduler: cosine max_steps: 200 micro_batch_size: 8 mixed_precision: bf16 mlflow_experiment_name: /tmp/ebaa36ac6b1bdb65_train_data.json model_type: AutoModelForCausalLM num_epochs: 1 optimizer: adamw_bnb_8bit output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false saves_per_epoch: 1 sequence_len: 1024 strict: false tf32: false tokenizer_type: AutoTokenizer train_on_inputs: false trust_remote_code: true val_set_size: 0.05 wandb_entity: null wandb_mode: online wandb_name: 33f6b38d-f8bd-4301-b3c9-673be809902f wandb_project: s56-4 wandb_run: your_name wandb_runid: 33f6b38d-f8bd-4301-b3c9-673be809902f warmup_steps: 5 weight_decay: 0.01 xformers_attention: true ``` </details><br> # ef4b672b-57ca-4395-820e-47f5c7dc8be4 This model is a fine-tuned version of [sethuiyer/Medichat-Llama3-8B](https://huggingface.co/sethuiyer/Medichat-Llama3-8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.8983 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-06 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 5 - training_steps: 200 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:------:\|:----:\|:---------------:\| \| 0.9723 \| 0.0802 \| 200 \| 0.8983 \| ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1
dinserazeebzxcsdfcvsd/fvbdfgbv	dinserazeebzxcsdfcvsd	2025-05-02T09:12:00Z	0	0	null	[ "license:creativeml-openrail-m", "region:us" ]	null	2025-05-02T09:12:00Z	--- license: creativeml-openrail-m ---

Subsets and Splits