llama_2_7b_MetaMathQA_40K_reverse
This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5038
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: 0.0001
- 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_steps: 0.02
- num_epochs: 1
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.8359 | 0.0211 | 13 | 0.6875 |
0.6478 | 0.0421 | 26 | 0.6519 |
0.6205 | 0.0632 | 39 | 0.6314 |
0.5969 | 0.0842 | 52 | 0.6175 |
0.5854 | 0.1053 | 65 | 0.6087 |
0.5693 | 0.1264 | 78 | 0.6002 |
0.577 | 0.1474 | 91 | 0.5936 |
0.5566 | 0.1685 | 104 | 0.5874 |
0.5817 | 0.1896 | 117 | 0.5830 |
0.5671 | 0.2106 | 130 | 0.5772 |
0.5488 | 0.2317 | 143 | 0.5731 |
0.5326 | 0.2527 | 156 | 0.5700 |
0.535 | 0.2738 | 169 | 0.5648 |
0.5342 | 0.2949 | 182 | 0.5593 |
0.5227 | 0.3159 | 195 | 0.5570 |
0.5319 | 0.3370 | 208 | 0.5515 |
0.5344 | 0.3580 | 221 | 0.5486 |
0.5282 | 0.3791 | 234 | 0.5451 |
0.5144 | 0.4002 | 247 | 0.5423 |
0.5182 | 0.4212 | 260 | 0.5402 |
0.5306 | 0.4423 | 273 | 0.5361 |
0.5152 | 0.4633 | 286 | 0.5349 |
0.5216 | 0.4844 | 299 | 0.5330 |
0.5222 | 0.5055 | 312 | 0.5290 |
0.5043 | 0.5265 | 325 | 0.5285 |
0.512 | 0.5476 | 338 | 0.5257 |
0.4953 | 0.5687 | 351 | 0.5240 |
0.5067 | 0.5897 | 364 | 0.5219 |
0.4957 | 0.6108 | 377 | 0.5210 |
0.5002 | 0.6318 | 390 | 0.5182 |
0.5028 | 0.6529 | 403 | 0.5170 |
0.5049 | 0.6740 | 416 | 0.5151 |
0.4981 | 0.6950 | 429 | 0.5129 |
0.4984 | 0.7161 | 442 | 0.5118 |
0.4901 | 0.7371 | 455 | 0.5100 |
0.4825 | 0.7582 | 468 | 0.5087 |
0.4981 | 0.7793 | 481 | 0.5082 |
0.4846 | 0.8003 | 494 | 0.5069 |
0.4774 | 0.8214 | 507 | 0.5059 |
0.4992 | 0.8424 | 520 | 0.5055 |
0.4948 | 0.8635 | 533 | 0.5050 |
0.4875 | 0.8846 | 546 | 0.5044 |
0.4889 | 0.9056 | 559 | 0.5040 |
0.4838 | 0.9267 | 572 | 0.5039 |
0.4891 | 0.9478 | 585 | 0.5039 |
0.485 | 0.9688 | 598 | 0.5039 |
0.4891 | 0.9899 | 611 | 0.5038 |
Framework versions
- PEFT 0.7.1
- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 0
Model tree for imdatta0/llama_2_7b_MetaMathQA_40K_reverse
Base model
meta-llama/Llama-2-7b-hf