--- license: mit base_model: microsoft/phi-2 tags: - generated_from_trainer model-index: - name: V0507HMA15HV1 results: [] --- # V0507HMA15HV1 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: 0.0614 ## 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: 100 - num_epochs: 3 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 1.9138 | 0.09 | 10 | 0.6327 | | 0.2601 | 0.18 | 20 | 0.1192 | | 0.1185 | 0.27 | 30 | 0.0927 | | 0.0993 | 0.36 | 40 | 0.0889 | | 0.0875 | 0.45 | 50 | 0.0865 | | 0.0947 | 0.54 | 60 | 0.0799 | | 0.0838 | 0.63 | 70 | 0.0754 | | 0.081 | 0.73 | 80 | 0.0781 | | 0.0859 | 0.82 | 90 | 0.0732 | | 0.0859 | 0.91 | 100 | 0.0735 | | 0.0786 | 1.0 | 110 | 0.0719 | | 0.0688 | 1.09 | 120 | 0.0721 | | 0.0738 | 1.18 | 130 | 0.0847 | | 0.0773 | 1.27 | 140 | 0.0688 | | 0.0674 | 1.36 | 150 | 0.0750 | | 0.0747 | 1.45 | 160 | 0.0687 | | 0.0675 | 1.54 | 170 | 0.0638 | | 0.0705 | 1.63 | 180 | 0.0662 | | 0.0632 | 1.72 | 190 | 0.0662 | | 0.0733 | 1.81 | 200 | 0.0658 | | 0.0569 | 1.9 | 210 | 0.0674 | | 0.0591 | 1.99 | 220 | 0.0592 | | 0.0434 | 2.08 | 230 | 0.0627 | | 0.0448 | 2.18 | 240 | 0.0648 | | 0.0423 | 2.27 | 250 | 0.0619 | | 0.0352 | 2.36 | 260 | 0.0664 | | 0.0447 | 2.45 | 270 | 0.0626 | | 0.0366 | 2.54 | 280 | 0.0621 | | 0.0359 | 2.63 | 290 | 0.0626 | | 0.0388 | 2.72 | 300 | 0.0631 | | 0.0403 | 2.81 | 310 | 0.0617 | | 0.0374 | 2.9 | 320 | 0.0615 | | 0.0382 | 2.99 | 330 | 0.0614 | ### Framework versions - Transformers 4.36.0.dev0 - Pytorch 2.1.2+cu121 - Datasets 2.18.0 - Tokenizers 0.14.1