Qwen2-7B_metamath_default
This model is a fine-tuned version of unsloth/Qwen2-7B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2142
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 | 0.1877 |
0.2034 | 0.0421 | 26 | 0.2530 |
0.261 | 0.0632 | 39 | 0.2844 |
0.2864 | 0.0842 | 52 | 0.3107 |
0.2984 | 0.1053 | 65 | 0.3264 |
0.318 | 0.1264 | 78 | 0.3361 |
0.3257 | 0.1474 | 91 | 0.3400 |
0.3139 | 0.1685 | 104 | 0.3377 |
0.327 | 0.1896 | 117 | 0.3429 |
0.3219 | 0.2106 | 130 | 0.3458 |
0.3325 | 0.2317 | 143 | 0.3458 |
0.3273 | 0.2527 | 156 | 0.3407 |
0.3328 | 0.2738 | 169 | 0.3395 |
0.32 | 0.2949 | 182 | 0.3372 |
0.3293 | 0.3159 | 195 | 0.3361 |
0.3182 | 0.3370 | 208 | 0.3306 |
0.3148 | 0.3580 | 221 | 0.3288 |
0.3128 | 0.3791 | 234 | 0.3255 |
0.3086 | 0.4002 | 247 | 0.3193 |
0.306 | 0.4212 | 260 | 0.3143 |
0.3122 | 0.4423 | 273 | 0.3120 |
0.2982 | 0.4633 | 286 | 0.3059 |
0.3023 | 0.4844 | 299 | 0.2964 |
0.2827 | 0.5055 | 312 | 0.2940 |
0.2844 | 0.5265 | 325 | 0.2864 |
0.2733 | 0.5476 | 338 | 0.2828 |
0.2711 | 0.5687 | 351 | 0.2777 |
0.2714 | 0.5897 | 364 | 0.2723 |
0.2605 | 0.6108 | 377 | 0.2652 |
0.2527 | 0.6318 | 390 | 0.2588 |
0.2526 | 0.6529 | 403 | 0.2529 |
0.2522 | 0.6740 | 416 | 0.2490 |
0.242 | 0.6950 | 429 | 0.2447 |
0.2443 | 0.7161 | 442 | 0.2400 |
0.2273 | 0.7371 | 455 | 0.2342 |
0.2288 | 0.7582 | 468 | 0.2318 |
0.2325 | 0.7793 | 481 | 0.2293 |
0.2264 | 0.8003 | 494 | 0.2262 |
0.2127 | 0.8214 | 507 | 0.2222 |
0.2166 | 0.8424 | 520 | 0.2199 |
0.2145 | 0.8635 | 533 | 0.2183 |
0.2089 | 0.8846 | 546 | 0.2167 |
0.206 | 0.9056 | 559 | 0.2158 |
0.2078 | 0.9267 | 572 | 0.2153 |
0.2117 | 0.9478 | 585 | 0.2147 |
0.2125 | 0.9688 | 598 | 0.2144 |
0.2118 | 0.9899 | 611 | 0.2142 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.4.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 52
Model tree for imdatta0/Qwen2-7B_metamath_default
Base model
unsloth/Qwen2-7B