metadata
library_name: peft
license: apache-2.0
base_model: mistralai/Mistral-7B-v0.1
tags:
- generated_from_trainer
model-index:
- name: mistral-mentalchat16k
results: []
mistral-mentalchat16k
This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6628
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7354 | 0.1496 | 100 | 0.7426 |
| 0.7341 | 0.2992 | 200 | 0.7197 |
| 0.6979 | 0.4488 | 300 | 0.7173 |
| 0.7061 | 0.5984 | 400 | 0.7045 |
| 0.707 | 0.7479 | 500 | 0.7023 |
| 0.6913 | 0.8975 | 600 | 0.6921 |
| 0.5442 | 1.0471 | 700 | 0.7099 |
| 0.4905 | 1.1967 | 800 | 0.7065 |
| 0.4704 | 1.3463 | 900 | 0.6926 |
| 0.4882 | 1.4959 | 1000 | 0.6873 |
| 0.4817 | 1.6455 | 1100 | 0.6804 |
| 0.4671 | 1.7951 | 1200 | 0.6698 |
| 0.4782 | 1.9447 | 1300 | 0.6681 |
| 0.2478 | 2.0942 | 1400 | 0.7225 |
| 0.2469 | 2.2438 | 1500 | 0.7441 |
| 0.2486 | 2.3934 | 1600 | 0.7288 |
| 0.2505 | 2.5430 | 1700 | 0.7424 |
| 0.2398 | 2.6926 | 1800 | 0.7298 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.3
- Pytorch 2.4.1+cu118
- Datasets 3.1.0
- Tokenizers 0.20.3