| --- |
| license: apache-2.0 |
| library_name: peft |
| tags: |
| - trl |
| - sft |
| - generated_from_trainer |
| datasets: |
| - generator |
| base_model: mistralai/Mistral-7B-Instruct-v0.2 |
| model-index: |
| - name: mistral7binstruct_summarize |
| 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. --> |
|
|
| # mistral7binstruct_summarize |
| |
| This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 1.5086 |
| |
| ## 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: 0.0005 |
| - 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 |
| - lr_scheduler_warmup_steps: 3 |
| - training_steps: 400 |
|
|
| ### Training results |
|
|
| | Training Loss | Epoch | Step | Validation Loss | |
| |:-------------:|:-----:|:----:|:---------------:| |
| | 1.7282 | 0.03 | 25 | 1.5986 | |
| | 1.5968 | 0.05 | 50 | 1.5566 | |
| | 1.5774 | 0.08 | 75 | 1.5433 | |
| | 1.5917 | 0.11 | 100 | 1.5409 | |
| | 1.5659 | 0.13 | 125 | 1.5339 | |
| | 1.5191 | 0.16 | 150 | 1.5312 | |
| | 1.5592 | 0.19 | 175 | 1.5267 | |
| | 1.4833 | 0.22 | 200 | 1.5245 | |
| | 1.4792 | 0.24 | 225 | 1.5208 | |
| | 1.5253 | 0.27 | 250 | 1.5162 | |
| | 1.5347 | 0.3 | 275 | 1.5138 | |
| | 1.5042 | 0.32 | 300 | 1.5116 | |
| | 1.4756 | 0.35 | 325 | 1.5098 | |
| | 1.5102 | 0.38 | 350 | 1.5090 | |
| | 1.4511 | 0.4 | 375 | 1.5087 | |
| | 1.4451 | 0.43 | 400 | 1.5086 | |
|
|
|
|
| ### Framework versions |
|
|
| - PEFT 0.9.0 |
| - Transformers 4.38.2 |
| - Pytorch 2.1.0+cu121 |
| - Datasets 2.18.0 |
| - Tokenizers 0.15.2 |
