|
--- |
|
base_model: unsloth/Qwen2-7B |
|
library_name: peft |
|
license: apache-2.0 |
|
tags: |
|
- unsloth |
|
- generated_from_trainer |
|
model-index: |
|
- name: Qwen2-7B_metamath_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. --> |
|
|
|
# Qwen2-7B_metamath_reverse |
|
|
|
This model is a fine-tuned version of [unsloth/Qwen2-7B](https://huggingface.co/unsloth/Qwen2-7B) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2136 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.1753 | 0.0211 | 13 | 0.1875 | |
|
| 0.2036 | 0.0421 | 26 | 0.2524 | |
|
| 0.2585 | 0.0632 | 39 | 0.2876 | |
|
| 0.2848 | 0.0842 | 52 | 0.3146 | |
|
| 0.2997 | 0.1053 | 65 | 0.3231 | |
|
| 0.3196 | 0.1264 | 78 | 0.3350 | |
|
| 0.3263 | 0.1474 | 91 | 0.3406 | |
|
| 0.3148 | 0.1685 | 104 | 0.3401 | |
|
| 0.3297 | 0.1896 | 117 | 0.3456 | |
|
| 0.3221 | 0.2106 | 130 | 0.3477 | |
|
| 0.3359 | 0.2317 | 143 | 0.3491 | |
|
| 0.3296 | 0.2527 | 156 | 0.3399 | |
|
| 0.3361 | 0.2738 | 169 | 0.3416 | |
|
| 0.3187 | 0.2949 | 182 | 0.3376 | |
|
| 0.3285 | 0.3159 | 195 | 0.3370 | |
|
| 0.3189 | 0.3370 | 208 | 0.3306 | |
|
| 0.3154 | 0.3580 | 221 | 0.3293 | |
|
| 0.3149 | 0.3791 | 234 | 0.3263 | |
|
| 0.3099 | 0.4002 | 247 | 0.3208 | |
|
| 0.3089 | 0.4212 | 260 | 0.3143 | |
|
| 0.3125 | 0.4423 | 273 | 0.3104 | |
|
| 0.2959 | 0.4633 | 286 | 0.3061 | |
|
| 0.3042 | 0.4844 | 299 | 0.2993 | |
|
| 0.2829 | 0.5055 | 312 | 0.2940 | |
|
| 0.2832 | 0.5265 | 325 | 0.2878 | |
|
| 0.2715 | 0.5476 | 338 | 0.2821 | |
|
| 0.2702 | 0.5687 | 351 | 0.2753 | |
|
| 0.2687 | 0.5897 | 364 | 0.2687 | |
|
| 0.2604 | 0.6108 | 377 | 0.2629 | |
|
| 0.252 | 0.6318 | 390 | 0.2579 | |
|
| 0.2537 | 0.6529 | 403 | 0.2529 | |
|
| 0.2535 | 0.6740 | 416 | 0.2477 | |
|
| 0.2442 | 0.6950 | 429 | 0.2425 | |
|
| 0.2451 | 0.7161 | 442 | 0.2378 | |
|
| 0.2275 | 0.7371 | 455 | 0.2338 | |
|
| 0.2288 | 0.7582 | 468 | 0.2310 | |
|
| 0.2323 | 0.7793 | 481 | 0.2294 | |
|
| 0.2254 | 0.8003 | 494 | 0.2260 | |
|
| 0.2142 | 0.8214 | 507 | 0.2221 | |
|
| 0.219 | 0.8424 | 520 | 0.2195 | |
|
| 0.2133 | 0.8635 | 533 | 0.2180 | |
|
| 0.2095 | 0.8846 | 546 | 0.2164 | |
|
| 0.2067 | 0.9056 | 559 | 0.2155 | |
|
| 0.2073 | 0.9267 | 572 | 0.2146 | |
|
| 0.2124 | 0.9478 | 585 | 0.2140 | |
|
| 0.2115 | 0.9688 | 598 | 0.2138 | |
|
| 0.2127 | 0.9899 | 611 | 0.2136 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.0 |
|
- Transformers 4.44.0 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
