--- license: mit base_model: microsoft/Phi-3-mini-4k-instruct tags: - generated_from_trainer model-index: - name: PHI30512HMAB3H results: [] --- # PHI30512HMAB3H This model is a fine-tuned version of [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.0270 ## 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: 16 - total_train_batch_size: 128 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: cosine_with_restarts - lr_scheduler_warmup_steps: 80 - num_epochs: 3 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 4.997 | 0.09 | 10 | 1.9314 | | 0.8757 | 0.18 | 20 | 0.2981 | | 0.313 | 0.27 | 30 | 0.2856 | | 0.2921 | 0.36 | 40 | 0.2521 | | 0.2598 | 0.45 | 50 | 0.2293 | | 0.237 | 0.54 | 60 | 0.2495 | | 0.2258 | 0.63 | 70 | 0.2233 | | 0.2146 | 0.73 | 80 | 0.1941 | | 0.2096 | 0.82 | 90 | 0.2012 | | 0.1865 | 0.91 | 100 | 0.1695 | | 0.1751 | 1.0 | 110 | 0.1659 | | 0.1633 | 1.09 | 120 | 0.1674 | | 0.17 | 1.18 | 130 | 0.1656 | | 0.1734 | 1.27 | 140 | 0.1610 | | 0.1633 | 1.36 | 150 | 0.1478 | | 0.1548 | 1.45 | 160 | 0.1397 | | 0.1564 | 1.54 | 170 | 0.1275 | | 0.1377 | 1.63 | 180 | 0.1395 | | 0.1093 | 1.72 | 190 | 0.0882 | | 0.1058 | 1.81 | 200 | 0.0842 | | 0.0908 | 1.9 | 210 | 0.0833 | | 0.0662 | 1.99 | 220 | 0.0539 | | 0.1051 | 2.08 | 230 | 0.1356 | | 0.1601 | 2.18 | 240 | 0.1337 | | 0.1836 | 2.27 | 250 | 0.0889 | | 0.067 | 2.36 | 260 | 0.0608 | | 0.0626 | 2.45 | 270 | 0.0509 | | 0.0477 | 2.54 | 280 | 0.0431 | | 0.0411 | 2.63 | 290 | 0.0390 | | 0.0349 | 2.72 | 300 | 0.0331 | | 0.0338 | 2.81 | 310 | 0.0298 | | 0.0288 | 2.9 | 320 | 0.0284 | | 0.0302 | 2.99 | 330 | 0.0270 | ### Framework versions - Transformers 4.36.0.dev0 - Pytorch 2.1.2+cu121 - Datasets 2.14.6 - Tokenizers 0.14.0