metadata
license: apache-2.0
base_model: mistralai/Mistral-7B-Instruct-v0.1
tags:
- generated_from_trainer
model-index:
- name: legal-mistral
results: []
legal-mistral
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: 2.3540
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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.4243 | 0.17 | 10 | 2.5809 |
| 2.32 | 0.34 | 20 | 2.5187 |
| 2.2652 | 0.52 | 30 | 2.4796 |
| 2.2424 | 0.69 | 40 | 2.4498 |
| 2.1759 | 0.86 | 50 | 2.4304 |
| 2.1698 | 1.03 | 60 | 2.4161 |
| 2.1337 | 1.21 | 70 | 2.4035 |
| 2.1193 | 1.38 | 80 | 2.3962 |
| 2.0928 | 1.55 | 90 | 2.3852 |
| 2.1072 | 1.72 | 100 | 2.3763 |
| 2.0701 | 1.9 | 110 | 2.3709 |
| 2.0584 | 2.07 | 120 | 2.3670 |
| 2.0569 | 2.24 | 130 | 2.3638 |
| 2.0396 | 2.41 | 140 | 2.3619 |
| 2.0406 | 2.59 | 150 | 2.3584 |
| 2.0259 | 2.76 | 160 | 2.3552 |
| 2.0492 | 2.93 | 170 | 2.3540 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0