|
--- |
|
base_model: unsloth/qwen2-7b-bnb-4bit |
|
library_name: peft |
|
license: apache-2.0 |
|
tags: |
|
- unsloth |
|
- generated_from_trainer |
|
model-index: |
|
- name: Qwen2-7B_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. --> |
|
|
|
# Qwen2-7B_metamath_ortho |
|
|
|
This model is a fine-tuned version of [unsloth/qwen2-7b-bnb-4bit](https://huggingface.co./unsloth/qwen2-7b-bnb-4bit) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2209 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.2058 | 0.0211 | 13 | 0.2001 | |
|
| 0.2083 | 0.0421 | 26 | 0.2441 | |
|
| 0.247 | 0.0632 | 39 | 0.2670 | |
|
| 0.2629 | 0.0842 | 52 | 0.2797 | |
|
| 0.2692 | 0.1053 | 65 | 0.2921 | |
|
| 0.2867 | 0.1264 | 78 | 0.2992 | |
|
| 0.2939 | 0.1474 | 91 | 0.3022 | |
|
| 0.2839 | 0.1685 | 104 | 0.3037 | |
|
| 0.2922 | 0.1896 | 117 | 0.3071 | |
|
| 0.2937 | 0.2106 | 130 | 0.3114 | |
|
| 0.3016 | 0.2317 | 143 | 0.3096 | |
|
| 0.298 | 0.2527 | 156 | 0.3078 | |
|
| 0.3059 | 0.2738 | 169 | 0.3083 | |
|
| 0.2883 | 0.2949 | 182 | 0.3052 | |
|
| 0.3013 | 0.3159 | 195 | 0.3010 | |
|
| 0.2874 | 0.3370 | 208 | 0.2995 | |
|
| 0.286 | 0.3580 | 221 | 0.2979 | |
|
| 0.2877 | 0.3791 | 234 | 0.2981 | |
|
| 0.2858 | 0.4002 | 247 | 0.2921 | |
|
| 0.2853 | 0.4212 | 260 | 0.2906 | |
|
| 0.2887 | 0.4423 | 273 | 0.2882 | |
|
| 0.2772 | 0.4633 | 286 | 0.2837 | |
|
| 0.2867 | 0.4844 | 299 | 0.2799 | |
|
| 0.2659 | 0.5055 | 312 | 0.2776 | |
|
| 0.2688 | 0.5265 | 325 | 0.2746 | |
|
| 0.2564 | 0.5476 | 338 | 0.2690 | |
|
| 0.2587 | 0.5687 | 351 | 0.2657 | |
|
| 0.2615 | 0.5897 | 364 | 0.2619 | |
|
| 0.2517 | 0.6108 | 377 | 0.2576 | |
|
| 0.2453 | 0.6318 | 390 | 0.2516 | |
|
| 0.2473 | 0.6529 | 403 | 0.2475 | |
|
| 0.2494 | 0.6740 | 416 | 0.2444 | |
|
| 0.2397 | 0.6950 | 429 | 0.2405 | |
|
| 0.2438 | 0.7161 | 442 | 0.2369 | |
|
| 0.2288 | 0.7371 | 455 | 0.2344 | |
|
| 0.2293 | 0.7582 | 468 | 0.2319 | |
|
| 0.2372 | 0.7793 | 481 | 0.2310 | |
|
| 0.2296 | 0.8003 | 494 | 0.2284 | |
|
| 0.2163 | 0.8214 | 507 | 0.2263 | |
|
| 0.2249 | 0.8424 | 520 | 0.2241 | |
|
| 0.219 | 0.8635 | 533 | 0.2234 | |
|
| 0.2154 | 0.8846 | 546 | 0.2230 | |
|
| 0.2107 | 0.9056 | 559 | 0.2221 | |
|
| 0.2132 | 0.9267 | 572 | 0.2215 | |
|
| 0.2202 | 0.9478 | 585 | 0.2212 | |
|
| 0.2197 | 0.9688 | 598 | 0.2211 | |
|
| 0.2168 | 0.9899 | 611 | 0.2209 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.0 |
|
- Transformers 4.44.0 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
