|
--- |
|
base_model: unsloth/mistral-7b-v0.3 |
|
library_name: peft |
|
license: apache-2.0 |
|
tags: |
|
- unsloth |
|
- generated_from_trainer |
|
model-index: |
|
- name: Mistral-7B-v0.3_metamath_ortho |
|
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_ortho |
|
|
|
This model is a fine-tuned version of [unsloth/mistral-7b-v0.3](https://huggingface.co/unsloth/mistral-7b-v0.3) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 3.8319 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.7761 | 0.0211 | 13 | 0.8475 | |
|
| 5.7285 | 0.0421 | 26 | 7.1242 | |
|
| 6.6463 | 0.0632 | 39 | 6.4624 | |
|
| 6.3183 | 0.0842 | 52 | 6.2700 | |
|
| 6.3056 | 0.1053 | 65 | 6.3511 | |
|
| 6.2849 | 0.1264 | 78 | 6.2801 | |
|
| 6.2952 | 0.1474 | 91 | 6.3205 | |
|
| 6.2939 | 0.1685 | 104 | 6.3566 | |
|
| 6.2779 | 0.1896 | 117 | 6.2580 | |
|
| 6.087 | 0.2106 | 130 | 5.9797 | |
|
| 5.8495 | 0.2317 | 143 | 5.8683 | |
|
| 5.6782 | 0.2527 | 156 | 5.5177 | |
|
| 5.4335 | 0.2738 | 169 | 5.3885 | |
|
| 5.4451 | 0.2949 | 182 | 5.7948 | |
|
| 5.5833 | 0.3159 | 195 | 5.2887 | |
|
| 5.2684 | 0.3370 | 208 | 5.3036 | |
|
| 5.1159 | 0.3580 | 221 | 5.1110 | |
|
| 5.0046 | 0.3791 | 234 | 4.9806 | |
|
| 4.9134 | 0.4002 | 247 | 4.9382 | |
|
| 4.9145 | 0.4212 | 260 | 4.9544 | |
|
| 4.7976 | 0.4423 | 273 | 4.7954 | |
|
| 4.7328 | 0.4633 | 286 | 4.6897 | |
|
| 4.6799 | 0.4844 | 299 | 4.5793 | |
|
| 4.5047 | 0.5055 | 312 | 4.6603 | |
|
| 4.529 | 0.5265 | 325 | 4.4405 | |
|
| 4.3835 | 0.5476 | 338 | 4.3916 | |
|
| 4.4279 | 0.5687 | 351 | 4.2860 | |
|
| 4.3177 | 0.5897 | 364 | 4.3171 | |
|
| 4.39 | 0.6108 | 377 | 4.3272 | |
|
| 4.3138 | 0.6318 | 390 | 4.3753 | |
|
| 4.2269 | 0.6529 | 403 | 4.3339 | |
|
| 4.1075 | 0.6740 | 416 | 4.1693 | |
|
| 4.2285 | 0.6950 | 429 | 4.1187 | |
|
| 4.1297 | 0.7161 | 442 | 4.1251 | |
|
| 4.0021 | 0.7371 | 455 | 4.0365 | |
|
| 4.0089 | 0.7582 | 468 | 4.0025 | |
|
| 3.9458 | 0.7793 | 481 | 3.9924 | |
|
| 3.9405 | 0.8003 | 494 | 3.9254 | |
|
| 3.9594 | 0.8214 | 507 | 3.8890 | |
|
| 3.9056 | 0.8424 | 520 | 3.8774 | |
|
| 3.8639 | 0.8635 | 533 | 3.8758 | |
|
| 3.8543 | 0.8846 | 546 | 3.8680 | |
|
| 3.9097 | 0.9056 | 559 | 3.8502 | |
|
| 3.8503 | 0.9267 | 572 | 3.8287 | |
|
| 3.789 | 0.9478 | 585 | 3.8357 | |
|
| 3.7923 | 0.9688 | 598 | 3.8299 | |
|
| 3.8071 | 0.9899 | 611 | 3.8319 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.0 |
|
- Transformers 4.44.0 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
