|
--- |
|
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_reverse |
|
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_reverse |
|
|
|
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: 0.4421 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.7252 | 0.0211 | 13 | 0.6225 | |
|
| 0.5814 | 0.0421 | 26 | 0.5900 | |
|
| 0.5599 | 0.0632 | 39 | 0.5751 | |
|
| 0.548 | 0.0842 | 52 | 0.5708 | |
|
| 0.5371 | 0.1053 | 65 | 0.5625 | |
|
| 0.5347 | 0.1264 | 78 | 0.5600 | |
|
| 0.5232 | 0.1474 | 91 | 0.5529 | |
|
| 0.5382 | 0.1685 | 104 | 0.5478 | |
|
| 0.5178 | 0.1896 | 117 | 0.5482 | |
|
| 0.5272 | 0.2106 | 130 | 0.5423 | |
|
| 0.5135 | 0.2317 | 143 | 0.5397 | |
|
| 0.4943 | 0.2527 | 156 | 0.5321 | |
|
| 0.5012 | 0.2738 | 169 | 0.5323 | |
|
| 0.5077 | 0.2949 | 182 | 0.5300 | |
|
| 0.5031 | 0.3159 | 195 | 0.5233 | |
|
| 0.506 | 0.3370 | 208 | 0.5238 | |
|
| 0.4851 | 0.3580 | 221 | 0.5180 | |
|
| 0.4915 | 0.3791 | 234 | 0.5146 | |
|
| 0.4826 | 0.4002 | 247 | 0.5150 | |
|
| 0.4964 | 0.4212 | 260 | 0.5096 | |
|
| 0.4989 | 0.4423 | 273 | 0.5050 | |
|
| 0.4846 | 0.4633 | 286 | 0.5021 | |
|
| 0.4776 | 0.4844 | 299 | 0.5006 | |
|
| 0.4725 | 0.5055 | 312 | 0.4927 | |
|
| 0.4752 | 0.5265 | 325 | 0.4898 | |
|
| 0.4719 | 0.5476 | 338 | 0.4862 | |
|
| 0.4689 | 0.5687 | 351 | 0.4817 | |
|
| 0.4573 | 0.5897 | 364 | 0.4772 | |
|
| 0.4536 | 0.6108 | 377 | 0.4754 | |
|
| 0.4536 | 0.6318 | 390 | 0.4700 | |
|
| 0.4519 | 0.6529 | 403 | 0.4664 | |
|
| 0.4448 | 0.6740 | 416 | 0.4633 | |
|
| 0.4327 | 0.6950 | 429 | 0.4618 | |
|
| 0.4528 | 0.7161 | 442 | 0.4586 | |
|
| 0.4379 | 0.7371 | 455 | 0.4557 | |
|
| 0.4504 | 0.7582 | 468 | 0.4537 | |
|
| 0.4436 | 0.7793 | 481 | 0.4525 | |
|
| 0.4451 | 0.8003 | 494 | 0.4497 | |
|
| 0.435 | 0.8214 | 507 | 0.4482 | |
|
| 0.4247 | 0.8424 | 520 | 0.4466 | |
|
| 0.4295 | 0.8635 | 533 | 0.4455 | |
|
| 0.4204 | 0.8846 | 546 | 0.4444 | |
|
| 0.4381 | 0.9056 | 559 | 0.4433 | |
|
| 0.4355 | 0.9267 | 572 | 0.4430 | |
|
| 0.4234 | 0.9478 | 585 | 0.4424 | |
|
| 0.4261 | 0.9688 | 598 | 0.4421 | |
|
| 0.4266 | 0.9899 | 611 | 0.4421 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.7.1 |
|
- Transformers 4.40.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
