End of training

README.md

@@ -1,73 +1,80 @@
+---
+license: apache-2.0
+library_name: peft
+tags:
+- trl
+- sft
+- generated_from_trainer
+base_model: mistralai/Mistral-7B-Instruct-v0.2
+model-index:
+- name: mistral-7b-json-quizz-fine-tuned-trl
+  results: []
+---
 
-    ---
-    language:
-    - en
-    library_name: transformers
-    tags:
-    - gpt
-    - llm
-    - large language model
-    - trl
-    inference: false
-    thumbnail: https://h2o.ai/etc.clientlibs/h2o/clientlibs/clientlib-site/resources/images/favicon.ico
-    ---
-    # Model Card
-    ## Summary
-
-    This model was trained using [TRL SFTTrainer](https://huggingface.co/docs/trl/main/en/sft_trainer).
-    - Base model: [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)
-
-    ## Model Description
-    Fine tuning to produce json quizz stability:
-
-    This quiz:
-
-    ```
-    Question: Who is considered the father of artificial intelligence?
-
-    A: Alan Turing
-    B: John McCarthy
-    C: Elon Musk
-    D: Stephen Hawking
-    Correct Answer: B
-    ```
-
-    Should return this json quiz:
-    ```json
-    {
-        "params": {
-            "questions": [
-            {
-                "question": "Who is considered the father of artificial intelligence?",
-                "A": "Alan Turing",
-                "B": "John McCarthy",
-                "C": "Elon Musk",
-                "D": "Stephen Hawking",
-                "reponse": "B"
-            }
-            ]
-        }
-    }
-    ```
-
-    ## How to Use
-    Provide instructions on how to use the model.
-
-    ## Training Data
-    Describe the data you used to train the model.
-
-    ## Training Procedure
-    Describe the training process, including any important parameters.
-
-    ## Evaluation Results
-    Include evaluation results if available.
-
-    ## Limitations and Bias
-    Discuss any limitations and potential biases in your model.
-
-    ## Ethical Considerations
-    Include any ethical considerations here.
-
-    ## Citation
-    Provide a citation if applicable.
-    
+<!-- 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. -->
+
+# mistral-7b-json-quizz-fine-tuned-trl
+
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6757
+
+## 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: 1
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- training_steps: 50
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.1134        | 4.55  | 50   | 0.6757          |
+
+
+### Framework versions
+
+- Transformers 4.37.0.dev0
+- Pytorch 2.1.1+cu121
+- Datasets 2.15.0
+- Tokenizers 0.15.0
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+
+### Framework versions
+
+
+- PEFT 0.6.2