| | --- |
| | license: mit |
| | base_model: microsoft/phi-2 |
| | tags: |
| | - generated_from_trainer |
| | model-index: |
| | - name: V0417MADP3 |
| | 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. --> |
| |
|
| | # V0417MADP3 |
| |
|
| | 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: 0.1901 |
| |
|
| | ## 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 | |
| | |:-------------:|:-----:|:----:|:---------------:| |
| | | 8.5507 | 0.09 | 10 | 3.0786 | |
| | | 6.3727 | 0.18 | 20 | 2.6464 | |
| | | 3.4656 | 0.27 | 30 | 1.9120 | |
| | | 1.5044 | 0.36 | 40 | 1.1144 | |
| | | 0.581 | 0.45 | 50 | 0.7389 | |
| | | 0.3434 | 0.54 | 60 | 0.5960 | |
| | | 0.3386 | 0.63 | 70 | 0.5215 | |
| | | 0.2957 | 0.73 | 80 | 0.5323 | |
| | | 0.258 | 0.82 | 90 | 0.4773 | |
| | | 0.263 | 0.91 | 100 | 0.4986 | |
| | | 0.2584 | 1.0 | 110 | 0.4831 | |
| | | 0.2808 | 1.09 | 120 | 0.5051 | |
| | | 0.2978 | 1.18 | 130 | 0.4790 | |
| | | 0.2479 | 1.27 | 140 | 0.4456 | |
| | | 0.4023 | 1.36 | 150 | 0.4223 | |
| | | 0.21 | 1.45 | 160 | 0.2159 | |
| | | 0.1788 | 1.54 | 170 | 0.2052 | |
| | | 0.1786 | 1.63 | 180 | 0.2024 | |
| | | 0.1748 | 1.72 | 190 | 0.2013 | |
| | | 0.1718 | 1.81 | 200 | 0.2138 | |
| | | 0.176 | 1.9 | 210 | 0.2197 | |
| | | 0.173 | 1.99 | 220 | 0.2321 | |
| | | 0.1877 | 2.08 | 230 | 0.2317 | |
| | | 0.1732 | 2.18 | 240 | 0.2126 | |
| | | 0.1661 | 2.27 | 250 | 0.1958 | |
| | | 0.1668 | 2.36 | 260 | 0.1955 | |
| | | 0.1642 | 2.45 | 270 | 0.1957 | |
| | | 0.1612 | 2.54 | 280 | 0.1937 | |
| | | 0.1681 | 2.63 | 290 | 0.1910 | |
| | | 0.1622 | 2.72 | 300 | 0.1901 | |
| | | 0.1592 | 2.81 | 310 | 0.1898 | |
| | | 0.1657 | 2.9 | 320 | 0.1904 | |
| | | 0.1696 | 2.99 | 330 | 0.1901 | |
| |
|
| |
|
| | ### Framework versions |
| |
|
| | - Transformers 4.36.0.dev0 |
| | - Pytorch 2.2.2+cu121 |
| | - Datasets 2.18.0 |
| | - Tokenizers 0.14.1 |
| |
|
