| | --- |
| | license: mit |
| | base_model: microsoft/phi-2 |
| | tags: |
| | - generated_from_trainer |
| | model-index: |
| | - name: V0422MADP6C |
| | results: [] |
| | --- |
| | |
| | <!-- This model card has been generated automatically according to the information the Trainer had access to. You |
| | should probably proofread and complete it, then remove this comment. --> |
| |
|
| | # V0422MADP6C |
| |
|
| | 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: 0.0637 |
| |
|
| | ## 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: 60 |
| | - num_epochs: 3 |
| | - mixed_precision_training: Native AMP |
| |
|
| | ### Training results |
| |
|
| | | Training Loss | Epoch | Step | Validation Loss | |
| | |:-------------:|:-----:|:----:|:---------------:| |
| | | 5.2356 | 0.09 | 10 | 1.9434 | |
| | | 2.8956 | 0.18 | 20 | 0.1595 | |
| | | 0.6107 | 0.27 | 30 | 0.1437 | |
| | | 0.1936 | 0.36 | 40 | 0.1236 | |
| | | 0.1283 | 0.45 | 50 | 0.1001 | |
| | | 0.1141 | 0.54 | 60 | 0.0983 | |
| | | 0.1042 | 0.63 | 70 | 0.0888 | |
| | | 0.089 | 0.73 | 80 | 0.0854 | |
| | | 0.0922 | 0.82 | 90 | 0.0815 | |
| | | 0.0892 | 0.91 | 100 | 0.0750 | |
| | | 0.0853 | 1.0 | 110 | 0.0789 | |
| | | 0.0755 | 1.09 | 120 | 0.0722 | |
| | | 0.0795 | 1.18 | 130 | 0.0764 | |
| | | 0.0794 | 1.27 | 140 | 0.0783 | |
| | | 0.0711 | 1.36 | 150 | 0.0753 | |
| | | 0.0717 | 1.45 | 160 | 0.0720 | |
| | | 0.067 | 1.54 | 170 | 0.0739 | |
| | | 0.0688 | 1.63 | 180 | 0.0712 | |
| | | 0.0654 | 1.72 | 190 | 0.0699 | |
| | | 0.0694 | 1.81 | 200 | 0.0652 | |
| | | 0.0621 | 1.9 | 210 | 0.0680 | |
| | | 0.0661 | 1.99 | 220 | 0.0654 | |
| | | 0.0515 | 2.08 | 230 | 0.0617 | |
| | | 0.0513 | 2.18 | 240 | 0.0650 | |
| | | 0.0462 | 2.27 | 250 | 0.0725 | |
| | | 0.0491 | 2.36 | 260 | 0.0693 | |
| | | 0.0538 | 2.45 | 270 | 0.0697 | |
| | | 0.0507 | 2.54 | 280 | 0.0663 | |
| | | 0.0437 | 2.63 | 290 | 0.0642 | |
| | | 0.0489 | 2.72 | 300 | 0.0635 | |
| | | 0.0485 | 2.81 | 310 | 0.0637 | |
| | | 0.0456 | 2.9 | 320 | 0.0637 | |
| | | 0.0557 | 2.99 | 330 | 0.0637 | |
| |
|
| |
|
| | ### Framework versions |
| |
|
| | - Transformers 4.36.0.dev0 |
| | - Pytorch 2.2.2+cu121 |
| | - Datasets 2.18.0 |
| | - Tokenizers 0.14.1 |
| |
|
