llama_2_7b_MetaMathQA_40K_ortho
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: 0.5037
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 |
---|---|---|---|
0.8682 | 0.0211 | 13 | 0.6814 |
0.641 | 0.0421 | 26 | 0.6427 |
0.6129 | 0.0632 | 39 | 0.6230 |
0.5885 | 0.0842 | 52 | 0.6083 |
0.5771 | 0.1053 | 65 | 0.5977 |
0.561 | 0.1264 | 78 | 0.5913 |
0.57 | 0.1474 | 91 | 0.5846 |
0.5493 | 0.1685 | 104 | 0.5778 |
0.5722 | 0.1896 | 117 | 0.5732 |
0.5579 | 0.2106 | 130 | 0.5689 |
0.5372 | 0.2317 | 143 | 0.5660 |
0.5243 | 0.2527 | 156 | 0.5615 |
0.5273 | 0.2738 | 169 | 0.5567 |
0.5274 | 0.2949 | 182 | 0.5524 |
0.5151 | 0.3159 | 195 | 0.5510 |
0.5254 | 0.3370 | 208 | 0.5461 |
0.5271 | 0.3580 | 221 | 0.5425 |
0.5193 | 0.3791 | 234 | 0.5386 |
0.508 | 0.4002 | 247 | 0.5376 |
0.5121 | 0.4212 | 260 | 0.5340 |
0.5259 | 0.4423 | 273 | 0.5321 |
0.51 | 0.4633 | 286 | 0.5306 |
0.5144 | 0.4844 | 299 | 0.5280 |
0.5165 | 0.5055 | 312 | 0.5253 |
0.498 | 0.5265 | 325 | 0.5232 |
0.5065 | 0.5476 | 338 | 0.5217 |
0.4899 | 0.5687 | 351 | 0.5212 |
0.5022 | 0.5897 | 364 | 0.5188 |
0.4932 | 0.6108 | 377 | 0.5162 |
0.4956 | 0.6318 | 390 | 0.5146 |
0.4999 | 0.6529 | 403 | 0.5142 |
0.5034 | 0.6740 | 416 | 0.5127 |
0.4962 | 0.6950 | 429 | 0.5107 |
0.4947 | 0.7161 | 442 | 0.5101 |
0.4885 | 0.7371 | 455 | 0.5092 |
0.4811 | 0.7582 | 468 | 0.5073 |
0.4965 | 0.7793 | 481 | 0.5069 |
0.4823 | 0.8003 | 494 | 0.5059 |
0.4747 | 0.8214 | 507 | 0.5052 |
0.4974 | 0.8424 | 520 | 0.5051 |
0.4928 | 0.8635 | 533 | 0.5049 |
0.4859 | 0.8846 | 546 | 0.5044 |
0.4899 | 0.9056 | 559 | 0.5040 |
0.4827 | 0.9267 | 572 | 0.5039 |
0.49 | 0.9478 | 585 | 0.5039 |
0.4833 | 0.9688 | 598 | 0.5038 |
0.4877 | 0.9899 | 611 | 0.5037 |
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_ortho
Base model
meta-llama/Llama-2-7b-hf
Adapter
this model