|
--- |
|
base_model: unsloth/llama-3-8b |
|
library_name: peft |
|
license: llama3 |
|
tags: |
|
- unsloth |
|
- generated_from_trainer |
|
model-index: |
|
- name: Meta-Llama-3-8B_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. --> |
|
|
|
# Meta-Llama-3-8B_metamath_reverse |
|
|
|
This model is a fine-tuned version of [unsloth/llama-3-8b](https://huggingface.co/unsloth/llama-3-8b) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5067 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.8667 | 0.0211 | 13 | 0.7332 | |
|
| 0.7009 | 0.0421 | 26 | 0.7322 | |
|
| 0.7161 | 0.0632 | 39 | 0.7229 | |
|
| 0.6909 | 0.0842 | 52 | 0.7190 | |
|
| 0.6541 | 0.1053 | 65 | 0.7083 | |
|
| 0.6704 | 0.1264 | 78 | 0.7014 | |
|
| 0.6806 | 0.1474 | 91 | 0.6999 | |
|
| 0.6735 | 0.1685 | 104 | 0.6932 | |
|
| 0.6509 | 0.1896 | 117 | 0.6962 | |
|
| 0.6537 | 0.2106 | 130 | 0.6907 | |
|
| 0.6508 | 0.2317 | 143 | 0.6892 | |
|
| 0.6594 | 0.2527 | 156 | 0.6816 | |
|
| 0.6534 | 0.2738 | 169 | 0.6734 | |
|
| 0.6559 | 0.2949 | 182 | 0.6744 | |
|
| 0.6391 | 0.3159 | 195 | 0.6739 | |
|
| 0.6115 | 0.3370 | 208 | 0.6628 | |
|
| 0.6261 | 0.3580 | 221 | 0.6548 | |
|
| 0.6288 | 0.3791 | 234 | 0.6545 | |
|
| 0.6377 | 0.4002 | 247 | 0.6510 | |
|
| 0.6106 | 0.4212 | 260 | 0.6465 | |
|
| 0.6203 | 0.4423 | 273 | 0.6377 | |
|
| 0.6196 | 0.4633 | 286 | 0.6276 | |
|
| 0.6146 | 0.4844 | 299 | 0.6216 | |
|
| 0.5931 | 0.5055 | 312 | 0.6187 | |
|
| 0.5926 | 0.5265 | 325 | 0.6058 | |
|
| 0.5807 | 0.5476 | 338 | 0.6018 | |
|
| 0.5738 | 0.5687 | 351 | 0.5915 | |
|
| 0.5509 | 0.5897 | 364 | 0.5852 | |
|
| 0.5641 | 0.6108 | 377 | 0.5815 | |
|
| 0.5606 | 0.6318 | 390 | 0.5723 | |
|
| 0.5478 | 0.6529 | 403 | 0.5653 | |
|
| 0.5451 | 0.6740 | 416 | 0.5613 | |
|
| 0.5362 | 0.6950 | 429 | 0.5556 | |
|
| 0.5328 | 0.7161 | 442 | 0.5474 | |
|
| 0.5185 | 0.7371 | 455 | 0.5413 | |
|
| 0.5127 | 0.7582 | 468 | 0.5359 | |
|
| 0.5036 | 0.7793 | 481 | 0.5299 | |
|
| 0.4922 | 0.8003 | 494 | 0.5265 | |
|
| 0.5246 | 0.8214 | 507 | 0.5219 | |
|
| 0.5088 | 0.8424 | 520 | 0.5175 | |
|
| 0.4908 | 0.8635 | 533 | 0.5150 | |
|
| 0.5091 | 0.8846 | 546 | 0.5120 | |
|
| 0.4902 | 0.9056 | 559 | 0.5096 | |
|
| 0.4865 | 0.9267 | 572 | 0.5083 | |
|
| 0.5007 | 0.9478 | 585 | 0.5072 | |
|
| 0.5001 | 0.9688 | 598 | 0.5068 | |
|
| 0.4989 | 0.9899 | 611 | 0.5067 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.0 |
|
- Transformers 4.44.0 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
