--- license: mit base_model: microsoft/phi-2 tags: - generated_from_trainer model-index: - name: V0503HMA6H results: [] --- # V0503HMA6H This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.0789 ## 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 | |:-------------:|:-----:|:----:|:---------------:| | 1.6387 | 0.09 | 10 | 0.1732 | | 0.1418 | 0.18 | 20 | 0.1123 | | 0.1121 | 0.27 | 30 | 0.0965 | | 0.1028 | 0.36 | 40 | 0.0867 | | 0.0843 | 0.45 | 50 | 0.0755 | | 0.0873 | 0.54 | 60 | 0.0846 | | 0.0852 | 0.63 | 70 | 0.0728 | | 0.0829 | 0.73 | 80 | 0.1861 | | 1.6498 | 0.82 | 90 | 0.1034 | | 2.9198 | 0.91 | 100 | 2.6894 | | 4.4394 | 1.0 | 110 | 2.1642 | | 1.1969 | 1.09 | 120 | 0.5102 | | 0.4174 | 1.18 | 130 | 0.3222 | | 0.3113 | 1.27 | 140 | 0.2990 | | 0.2566 | 1.36 | 150 | 0.2072 | | 0.1857 | 1.45 | 160 | 0.1786 | | 0.1804 | 1.54 | 170 | 0.1617 | | 0.1623 | 1.63 | 180 | 0.1588 | | 0.1511 | 1.72 | 190 | 0.1394 | | 0.1353 | 1.81 | 200 | 0.1214 | | 0.1185 | 1.9 | 210 | 0.1228 | | 0.1201 | 1.99 | 220 | 0.1079 | | 0.1073 | 2.08 | 230 | 0.1116 | | 0.1001 | 2.18 | 240 | 0.0929 | | 0.0964 | 2.27 | 250 | 0.0884 | | 0.0897 | 2.36 | 260 | 0.0858 | | 0.0862 | 2.45 | 270 | 0.0836 | | 0.0866 | 2.54 | 280 | 0.0830 | | 0.086 | 2.63 | 290 | 0.0805 | | 0.0818 | 2.72 | 300 | 0.0786 | | 0.084 | 2.81 | 310 | 0.0774 | | 0.0802 | 2.9 | 320 | 0.0790 | | 0.0759 | 2.99 | 330 | 0.0789 | ### Framework versions - Transformers 4.36.0.dev0 - Pytorch 2.1.2+cu121 - Datasets 2.18.0 - Tokenizers 0.14.1