|
--- |
|
license: apache-2.0 |
|
library_name: peft |
|
tags: |
|
- unsloth |
|
- generated_from_trainer |
|
base_model: mistralai/Mistral-7B-v0.3 |
|
model-index: |
|
- name: mistral_7b_v_MetaMathQA_40K |
|
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_v_MetaMathQA_40K |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 4.0534 |
|
|
|
## 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: 0.0001 |
|
- 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_steps: 0.02 |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.8546 | 0.0211 | 13 | 9.0448 | |
|
| 8.7033 | 0.0421 | 26 | 6.8246 | |
|
| 7.1208 | 0.0632 | 39 | 6.6756 | |
|
| 6.5364 | 0.0842 | 52 | 6.5704 | |
|
| 6.4506 | 0.1053 | 65 | 6.4165 | |
|
| 6.3651 | 0.1264 | 78 | 6.4591 | |
|
| 6.4236 | 0.1474 | 91 | 6.3382 | |
|
| 6.3751 | 0.1685 | 104 | 6.3491 | |
|
| 6.29 | 0.1896 | 117 | 6.3231 | |
|
| 6.1703 | 0.2106 | 130 | 6.1876 | |
|
| 5.9486 | 0.2317 | 143 | 5.8240 | |
|
| 5.7357 | 0.2527 | 156 | 5.6677 | |
|
| 5.5395 | 0.2738 | 169 | 5.7816 | |
|
| 5.4509 | 0.2949 | 182 | 5.4254 | |
|
| 5.4296 | 0.3159 | 195 | 5.2703 | |
|
| 5.3284 | 0.3370 | 208 | 5.1638 | |
|
| 5.2125 | 0.3580 | 221 | 5.1691 | |
|
| 5.0807 | 0.3791 | 234 | 5.0448 | |
|
| 4.9527 | 0.4002 | 247 | 4.9290 | |
|
| 4.929 | 0.4212 | 260 | 4.9626 | |
|
| 4.9299 | 0.4423 | 273 | 4.8930 | |
|
| 4.8363 | 0.4633 | 286 | 4.6863 | |
|
| 4.6998 | 0.4844 | 299 | 4.6888 | |
|
| 4.6004 | 0.5055 | 312 | 4.6411 | |
|
| 4.6229 | 0.5265 | 325 | 4.5178 | |
|
| 4.4437 | 0.5476 | 338 | 4.4411 | |
|
| 4.4564 | 0.5687 | 351 | 4.4293 | |
|
| 4.4144 | 0.5897 | 364 | 4.3946 | |
|
| 4.3888 | 0.6108 | 377 | 4.3527 | |
|
| 4.3296 | 0.6318 | 390 | 4.2652 | |
|
| 4.2489 | 0.6529 | 403 | 4.2610 | |
|
| 4.2046 | 0.6740 | 416 | 4.2029 | |
|
| 4.2525 | 0.6950 | 429 | 4.1885 | |
|
| 4.2439 | 0.7161 | 442 | 4.1833 | |
|
| 4.141 | 0.7371 | 455 | 4.1576 | |
|
| 4.1417 | 0.7582 | 468 | 4.1388 | |
|
| 4.1334 | 0.7793 | 481 | 4.1094 | |
|
| 4.1319 | 0.8003 | 494 | 4.0910 | |
|
| 4.1122 | 0.8214 | 507 | 4.1114 | |
|
| 4.0976 | 0.8424 | 520 | 4.0905 | |
|
| 4.0836 | 0.8635 | 533 | 4.0963 | |
|
| 4.061 | 0.8846 | 546 | 4.0767 | |
|
| 4.1107 | 0.9056 | 559 | 4.0573 | |
|
| 4.0673 | 0.9267 | 572 | 4.0522 | |
|
| 4.0283 | 0.9478 | 585 | 4.0558 | |
|
| 4.045 | 0.9688 | 598 | 4.0532 | |
|
| 4.0369 | 0.9899 | 611 | 4.0534 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.7.1 |
|
- Transformers 4.40.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |