|
--- |
|
base_model: unsloth/gemma-2-9b |
|
library_name: peft |
|
license: gemma |
|
tags: |
|
- unsloth |
|
- generated_from_trainer |
|
model-index: |
|
- name: gemma-2-9b_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. --> |
|
|
|
# gemma-2-9b_metamath_ortho |
|
|
|
This model is a fine-tuned version of [unsloth/gemma-2-9b](https://huggingface.co./unsloth/gemma-2-9b) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 8.8949 |
|
|
|
## 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: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 32 |
|
- 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 | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.7552 | 0.0211 | 13 | 11.3000 | |
|
| 11.6101 | 0.0421 | 26 | 11.6207 | |
|
| 11.2097 | 0.0632 | 39 | 11.2115 | |
|
| 10.6416 | 0.0843 | 52 | 11.2629 | |
|
| 11.087 | 0.1053 | 65 | 11.0208 | |
|
| 10.911 | 0.1264 | 78 | 11.1743 | |
|
| 10.3864 | 0.1474 | 91 | 9.9812 | |
|
| 10.0221 | 0.1685 | 104 | 9.7955 | |
|
| 10.1644 | 0.1896 | 117 | 10.3842 | |
|
| 10.6605 | 0.2106 | 130 | 10.8225 | |
|
| 10.6448 | 0.2317 | 143 | 10.6978 | |
|
| 10.6678 | 0.2528 | 156 | 10.5545 | |
|
| 10.3694 | 0.2738 | 169 | 9.2694 | |
|
| 11.5024 | 0.2949 | 182 | 11.9171 | |
|
| 11.731 | 0.3159 | 195 | 11.1801 | |
|
| 11.5971 | 0.3370 | 208 | 10.8633 | |
|
| 11.0785 | 0.3581 | 221 | 10.8301 | |
|
| 10.5228 | 0.3791 | 234 | 10.5660 | |
|
| 10.4021 | 0.4002 | 247 | 10.2953 | |
|
| 10.2801 | 0.4213 | 260 | 10.1140 | |
|
| 11.5045 | 0.4423 | 273 | 11.9247 | |
|
| 11.7387 | 0.4634 | 286 | 11.5647 | |
|
| 11.6365 | 0.4845 | 299 | 11.5171 | |
|
| 10.8443 | 0.5055 | 312 | 9.7961 | |
|
| 11.0662 | 0.5266 | 325 | 10.9008 | |
|
| 10.9411 | 0.5476 | 338 | 11.0820 | |
|
| 11.2701 | 0.5687 | 351 | 11.4640 | |
|
| 10.8777 | 0.5898 | 364 | 9.6432 | |
|
| 10.3239 | 0.6108 | 377 | 10.6138 | |
|
| 10.8646 | 0.6319 | 390 | 10.7791 | |
|
| 11.2221 | 0.6530 | 403 | 11.1868 | |
|
| 11.0328 | 0.6740 | 416 | 10.8502 | |
|
| 10.8706 | 0.6951 | 429 | 10.7639 | |
|
| 10.9227 | 0.7162 | 442 | 10.7393 | |
|
| 10.3582 | 0.7372 | 455 | 8.8997 | |
|
| 9.4372 | 0.7583 | 468 | 9.9192 | |
|
| 9.5329 | 0.7793 | 481 | 9.3924 | |
|
| 9.6104 | 0.8004 | 494 | 9.3153 | |
|
| 9.3477 | 0.8215 | 507 | 9.2830 | |
|
| 9.7923 | 0.8425 | 520 | 8.8935 | |
|
| 8.8863 | 0.8636 | 533 | 9.5862 | |
|
| 8.9213 | 0.8847 | 546 | 8.4786 | |
|
| 8.5736 | 0.9057 | 559 | 9.8808 | |
|
| 9.5766 | 0.9268 | 572 | 8.6842 | |
|
| 8.6045 | 0.9478 | 585 | 8.7064 | |
|
| 8.6253 | 0.9689 | 598 | 8.8377 | |
|
| 8.8052 | 0.9900 | 611 | 8.8949 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.0 |
|
- Transformers 4.44.0 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
