legal-mistral-v2
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.1 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.9807
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: 80
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.9084 | 0.09 | 100 | 2.5259 |
| 2.5097 | 0.17 | 200 | 2.3141 |
| 2.377 | 0.26 | 300 | 2.1949 |
| 2.2432 | 0.34 | 400 | 2.1256 |
| 2.215 | 0.43 | 500 | 2.0810 |
| 2.1517 | 0.52 | 600 | 2.0496 |
| 2.0899 | 0.6 | 700 | 2.0256 |
| 2.0947 | 0.69 | 800 | 2.0088 |
| 2.0748 | 0.77 | 900 | 1.9961 |
| 2.0845 | 0.86 | 1000 | 1.9865 |
| 2.0586 | 0.95 | 1100 | 1.9807 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0