|
--- |
|
base_model: unsloth/mistral-7b-v0.3-bnb-4bit |
|
library_name: peft |
|
license: apache-2.0 |
|
tags: |
|
- unsloth |
|
- generated_from_trainer |
|
model-index: |
|
- name: Mistral-7B-v0.3_metamath_default |
|
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. --> |
|
|
|
# Mistral-7B-v0.3_metamath_default |
|
|
|
This model is a fine-tuned version of [unsloth/mistral-7b-v0.3-bnb-4bit](https://huggingface.co/unsloth/mistral-7b-v0.3-bnb-4bit) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 4.0734 |
|
|
|
## 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: 0.0003 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 64 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_ratio: 0.02 |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.7559 | 0.0211 | 13 | 8.3985 | |
|
| 9.4604 | 0.0421 | 26 | 6.8141 | |
|
| 6.7353 | 0.0632 | 39 | 6.4223 | |
|
| 6.4242 | 0.0842 | 52 | 6.2759 | |
|
| 6.1115 | 0.1053 | 65 | 6.0333 | |
|
| 5.9214 | 0.1264 | 78 | 5.8343 | |
|
| 5.6735 | 0.1474 | 91 | 5.5846 | |
|
| 5.557 | 0.1685 | 104 | 5.4916 | |
|
| 5.3297 | 0.1896 | 117 | 5.2345 | |
|
| 5.1963 | 0.2106 | 130 | 5.1310 | |
|
| 5.1252 | 0.2317 | 143 | 5.0674 | |
|
| 4.983 | 0.2527 | 156 | 4.9390 | |
|
| 4.8933 | 0.2738 | 169 | 4.8252 | |
|
| 4.7722 | 0.2949 | 182 | 4.7449 | |
|
| 4.7722 | 0.3159 | 195 | 4.7386 | |
|
| 4.6446 | 0.3370 | 208 | 4.6346 | |
|
| 4.5823 | 0.3580 | 221 | 4.5544 | |
|
| 4.576 | 0.3791 | 234 | 4.5238 | |
|
| 4.5056 | 0.4002 | 247 | 4.6538 | |
|
| 4.5501 | 0.4212 | 260 | 4.4766 | |
|
| 4.5197 | 0.4423 | 273 | 4.4369 | |
|
| 4.6259 | 0.4633 | 286 | 4.4561 | |
|
| 4.546 | 0.4844 | 299 | 4.4278 | |
|
| 4.3478 | 0.5055 | 312 | 4.3790 | |
|
| 4.3754 | 0.5265 | 325 | 4.3635 | |
|
| 4.2714 | 0.5476 | 338 | 4.3611 | |
|
| 4.3724 | 0.5687 | 351 | 4.3629 | |
|
| 4.2961 | 0.5897 | 364 | 4.2578 | |
|
| 4.2806 | 0.6108 | 377 | 4.2863 | |
|
| 4.3088 | 0.6318 | 390 | 4.2221 | |
|
| 4.2165 | 0.6529 | 403 | 4.2158 | |
|
| 4.1776 | 0.6740 | 416 | 4.1896 | |
|
| 4.2615 | 0.6950 | 429 | 4.3146 | |
|
| 4.2536 | 0.7161 | 442 | 4.2153 | |
|
| 4.1308 | 0.7371 | 455 | 4.1701 | |
|
| 4.1749 | 0.7582 | 468 | 4.1346 | |
|
| 4.1219 | 0.7793 | 481 | 4.1276 | |
|
| 4.136 | 0.8003 | 494 | 4.1162 | |
|
| 4.1453 | 0.8214 | 507 | 4.1070 | |
|
| 4.1025 | 0.8424 | 520 | 4.1167 | |
|
| 4.1207 | 0.8635 | 533 | 4.0925 | |
|
| 4.0847 | 0.8846 | 546 | 4.0926 | |
|
| 4.1504 | 0.9056 | 559 | 4.0795 | |
|
| 4.1211 | 0.9267 | 572 | 4.0711 | |
|
| 4.038 | 0.9478 | 585 | 4.0763 | |
|
| 4.0944 | 0.9688 | 598 | 4.0744 | |
|
| 4.0771 | 0.9899 | 611 | 4.0734 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.0 |
|
- Transformers 4.44.0 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |