|
--- |
|
base_model: unsloth/llama-3-8b |
|
library_name: peft |
|
license: llama3 |
|
tags: |
|
- unsloth |
|
- generated_from_trainer |
|
model-index: |
|
- name: Meta-Llama-3-8B_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. --> |
|
|
|
# Meta-Llama-3-8B_metamath_ortho |
|
|
|
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: 0.4760 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.9348 | 0.0211 | 13 | 0.7026 | |
|
| 0.6609 | 0.0421 | 26 | 0.6960 | |
|
| 0.6695 | 0.0632 | 39 | 0.6785 | |
|
| 0.6578 | 0.0842 | 52 | 0.6770 | |
|
| 0.6222 | 0.1053 | 65 | 0.6748 | |
|
| 0.6331 | 0.1264 | 78 | 0.6662 | |
|
| 0.6413 | 0.1474 | 91 | 0.6609 | |
|
| 0.631 | 0.1685 | 104 | 0.6538 | |
|
| 0.6115 | 0.1896 | 117 | 0.6490 | |
|
| 0.6097 | 0.2106 | 130 | 0.6458 | |
|
| 0.6052 | 0.2317 | 143 | 0.6533 | |
|
| 0.6131 | 0.2527 | 156 | 0.6335 | |
|
| 0.6012 | 0.2738 | 169 | 0.6165 | |
|
| 0.6002 | 0.2949 | 182 | 0.6204 | |
|
| 0.5849 | 0.3159 | 195 | 0.6220 | |
|
| 0.5689 | 0.3370 | 208 | 0.6159 | |
|
| 0.5781 | 0.3580 | 221 | 0.6110 | |
|
| 0.5765 | 0.3791 | 234 | 0.6027 | |
|
| 0.5899 | 0.4002 | 247 | 0.5983 | |
|
| 0.5638 | 0.4212 | 260 | 0.5905 | |
|
| 0.5716 | 0.4423 | 273 | 0.5874 | |
|
| 0.5729 | 0.4633 | 286 | 0.5809 | |
|
| 0.5691 | 0.4844 | 299 | 0.5729 | |
|
| 0.5441 | 0.5055 | 312 | 0.5659 | |
|
| 0.5468 | 0.5265 | 325 | 0.5584 | |
|
| 0.536 | 0.5476 | 338 | 0.5544 | |
|
| 0.5277 | 0.5687 | 351 | 0.5474 | |
|
| 0.5052 | 0.5897 | 364 | 0.5397 | |
|
| 0.5185 | 0.6108 | 377 | 0.5309 | |
|
| 0.5161 | 0.6318 | 390 | 0.5262 | |
|
| 0.5056 | 0.6529 | 403 | 0.5227 | |
|
| 0.5091 | 0.6740 | 416 | 0.5164 | |
|
| 0.492 | 0.6950 | 429 | 0.5115 | |
|
| 0.4936 | 0.7161 | 442 | 0.5070 | |
|
| 0.4818 | 0.7371 | 455 | 0.5005 | |
|
| 0.4762 | 0.7582 | 468 | 0.4986 | |
|
| 0.4685 | 0.7793 | 481 | 0.4938 | |
|
| 0.4614 | 0.8003 | 494 | 0.4904 | |
|
| 0.4942 | 0.8214 | 507 | 0.4870 | |
|
| 0.4767 | 0.8424 | 520 | 0.4837 | |
|
| 0.4589 | 0.8635 | 533 | 0.4819 | |
|
| 0.4806 | 0.8846 | 546 | 0.4796 | |
|
| 0.4647 | 0.9056 | 559 | 0.4782 | |
|
| 0.461 | 0.9267 | 572 | 0.4773 | |
|
| 0.4718 | 0.9478 | 585 | 0.4766 | |
|
| 0.4684 | 0.9688 | 598 | 0.4761 | |
|
| 0.4716 | 0.9899 | 611 | 0.4760 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.0 |
|
- Transformers 4.44.0 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |