|
--- |
|
license: mit |
|
base_model: microsoft/phi-2 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: V0503HMA5H |
|
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. --> |
|
|
|
# V0503HMA5H |
|
|
|
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: 0.1346 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.7787 | 0.09 | 10 | 0.1611 | |
|
| 0.1597 | 0.18 | 20 | 0.1230 | |
|
| 0.1177 | 0.27 | 30 | 0.1023 | |
|
| 0.1021 | 0.36 | 40 | 0.0896 | |
|
| 0.085 | 0.45 | 50 | 0.0808 | |
|
| 0.0884 | 0.54 | 60 | 0.0808 | |
|
| 0.0855 | 0.63 | 70 | 0.0706 | |
|
| 0.0789 | 0.73 | 80 | 0.0902 | |
|
| 0.087 | 0.82 | 90 | 0.0869 | |
|
| 0.1125 | 0.91 | 100 | 8.7126 | |
|
| 2.2018 | 1.0 | 110 | 0.4319 | |
|
| 0.2705 | 1.09 | 120 | 0.2003 | |
|
| 0.759 | 1.18 | 130 | 0.2586 | |
|
| 0.2778 | 1.27 | 140 | 0.1786 | |
|
| 0.191 | 1.36 | 150 | 0.2223 | |
|
| 0.177 | 1.45 | 160 | 0.1639 | |
|
| 0.1691 | 1.54 | 170 | 0.1591 | |
|
| 0.16 | 1.63 | 180 | 0.1638 | |
|
| 0.1535 | 1.72 | 190 | 0.1508 | |
|
| 0.1501 | 1.81 | 200 | 0.1572 | |
|
| 0.1549 | 1.9 | 210 | 0.1487 | |
|
| 0.1523 | 1.99 | 220 | 0.1505 | |
|
| 0.1538 | 2.08 | 230 | 0.1558 | |
|
| 0.1493 | 2.18 | 240 | 0.1474 | |
|
| 0.1438 | 2.27 | 250 | 0.1439 | |
|
| 0.1455 | 2.36 | 260 | 0.1425 | |
|
| 0.1406 | 2.45 | 270 | 0.1433 | |
|
| 0.1402 | 2.54 | 280 | 0.1382 | |
|
| 0.1371 | 2.63 | 290 | 0.1385 | |
|
| 0.138 | 2.72 | 300 | 0.1355 | |
|
| 0.1352 | 2.81 | 310 | 0.1354 | |
|
| 0.1366 | 2.9 | 320 | 0.1347 | |
|
| 0.1368 | 2.99 | 330 | 0.1346 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.14.1 |
|
|
