llama_2_7b_MetaMathQA_40K_ortho_scale15
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: 0.5338
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 |
---|---|---|---|
0.9029 | 0.0211 | 13 | 0.7003 |
0.6583 | 0.0421 | 26 | 0.6563 |
0.6261 | 0.0632 | 39 | 0.6368 |
0.6011 | 0.0842 | 52 | 0.6237 |
0.5912 | 0.1053 | 65 | 0.6134 |
0.5747 | 0.1264 | 78 | 0.6052 |
0.5835 | 0.1474 | 91 | 0.5995 |
0.5632 | 0.1685 | 104 | 0.5931 |
0.5872 | 0.1896 | 117 | 0.5894 |
0.5735 | 0.2106 | 130 | 0.5854 |
0.5546 | 0.2317 | 143 | 0.5818 |
0.5403 | 0.2527 | 156 | 0.5778 |
0.5436 | 0.2738 | 169 | 0.5742 |
0.5416 | 0.2949 | 182 | 0.5713 |
0.5298 | 0.3159 | 195 | 0.5691 |
0.5431 | 0.3370 | 208 | 0.5658 |
0.5443 | 0.3580 | 221 | 0.5629 |
0.537 | 0.3791 | 234 | 0.5595 |
0.5258 | 0.4002 | 247 | 0.5582 |
0.5309 | 0.4212 | 260 | 0.5557 |
0.546 | 0.4423 | 273 | 0.5538 |
0.5299 | 0.4633 | 286 | 0.5519 |
0.5347 | 0.4844 | 299 | 0.5502 |
0.5372 | 0.5055 | 312 | 0.5486 |
0.5197 | 0.5265 | 325 | 0.5473 |
0.5269 | 0.5476 | 338 | 0.5458 |
0.5135 | 0.5687 | 351 | 0.5453 |
0.5217 | 0.5897 | 364 | 0.5435 |
0.5171 | 0.6108 | 377 | 0.5423 |
0.5184 | 0.6318 | 390 | 0.5413 |
0.5246 | 0.6529 | 403 | 0.5406 |
0.5258 | 0.6740 | 416 | 0.5397 |
0.5216 | 0.6950 | 429 | 0.5383 |
0.5201 | 0.7161 | 442 | 0.5376 |
0.514 | 0.7371 | 455 | 0.5374 |
0.5085 | 0.7582 | 468 | 0.5363 |
0.5217 | 0.7793 | 481 | 0.5358 |
0.5089 | 0.8003 | 494 | 0.5350 |
0.5021 | 0.8214 | 507 | 0.5345 |
0.5238 | 0.8424 | 520 | 0.5344 |
0.5213 | 0.8635 | 533 | 0.5345 |
0.5142 | 0.8846 | 546 | 0.5342 |
0.5152 | 0.9056 | 559 | 0.5340 |
0.5089 | 0.9267 | 572 | 0.5338 |
0.5165 | 0.9478 | 585 | 0.5338 |
0.5128 | 0.9688 | 598 | 0.5337 |
0.5143 | 0.9899 | 611 | 0.5338 |
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
- 2
Model tree for imdatta0/llama_2_7b_MetaMathQA_40K_ortho_scale15
Base model
meta-llama/Llama-2-7b-hf
Adapter
this model