| | --- |
| | license: mit |
| | base_model: microsoft/phi-2 |
| | tags: |
| | - generated_from_trainer |
| | model-index: |
| | - name: V0424MADP3 |
| | 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. --> |
| |
|
| | # V0424MADP3 |
| |
|
| | 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: 0.1501 |
| |
|
| | ## 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 | |
| | |:-------------:|:-----:|:----:|:---------------:| |
| | | 8.3928 | 0.09 | 10 | 2.9739 | |
| | | 4.5311 | 0.18 | 20 | 1.7154 | |
| | | 1.0104 | 0.27 | 30 | 0.4752 | |
| | | 0.2358 | 0.36 | 40 | 0.2687 | |
| | | 0.1727 | 0.45 | 50 | 0.1997 | |
| | | 0.1674 | 0.54 | 60 | 0.1636 | |
| | | 0.1595 | 0.63 | 70 | 0.1496 | |
| | | 0.1574 | 0.73 | 80 | 0.1540 | |
| | | 0.1562 | 0.82 | 90 | 0.1556 | |
| | | 0.1541 | 0.91 | 100 | 0.1491 | |
| | | 0.1635 | 1.0 | 110 | 0.1578 | |
| | | 0.155 | 1.09 | 120 | 0.1801 | |
| | | 0.167 | 1.18 | 130 | 0.1961 | |
| | | 0.1638 | 1.27 | 140 | 0.1946 | |
| | | 0.165 | 1.36 | 150 | 0.1727 | |
| | | 0.1548 | 1.45 | 160 | 0.1608 | |
| | | 0.1581 | 1.54 | 170 | 0.1598 | |
| | | 0.1529 | 1.63 | 180 | 0.1563 | |
| | | 0.1539 | 1.72 | 190 | 0.1543 | |
| | | 0.1555 | 1.81 | 200 | 0.1554 | |
| | | 0.1592 | 1.9 | 210 | 0.1527 | |
| | | 0.1553 | 1.99 | 220 | 0.1647 | |
| | | 0.1584 | 2.08 | 230 | 0.1608 | |
| | | 0.1515 | 2.18 | 240 | 0.1573 | |
| | | 0.148 | 2.27 | 250 | 0.1511 | |
| | | 0.1503 | 2.36 | 260 | 0.1536 | |
| | | 0.15 | 2.45 | 270 | 0.1516 | |
| | | 0.1476 | 2.54 | 280 | 0.1491 | |
| | | 0.1473 | 2.63 | 290 | 0.1486 | |
| | | 0.1472 | 2.72 | 300 | 0.1501 | |
| | | 0.1468 | 2.81 | 310 | 0.1501 | |
| | | 0.1489 | 2.9 | 320 | 0.1501 | |
| | | 0.149 | 2.99 | 330 | 0.1501 | |
| |
|
| |
|
| | ### Framework versions |
| |
|
| | - Transformers 4.36.0.dev0 |
| | - Pytorch 2.1.2+cu121 |
| | - Datasets 2.18.0 |
| | - Tokenizers 0.14.1 |
| |
|
