|
--- |
|
license: mit |
|
base_model: microsoft/phi-2 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: V0415MA3plus |
|
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. --> |
|
|
|
# V0415MA3plus |
|
|
|
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: 0.0764 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.7971 | 0.09 | 10 | 0.1919 | |
|
| 0.1458 | 0.18 | 20 | 0.1068 | |
|
| 0.1006 | 0.27 | 30 | 0.0813 | |
|
| 0.0868 | 0.36 | 40 | 0.0727 | |
|
| 0.0776 | 0.45 | 50 | 0.0726 | |
|
| 0.083 | 0.54 | 60 | 0.0739 | |
|
| 0.0688 | 0.63 | 70 | 0.0666 | |
|
| 0.0645 | 0.73 | 80 | 0.0682 | |
|
| 0.0721 | 0.82 | 90 | 0.0621 | |
|
| 0.0742 | 0.91 | 100 | 0.0638 | |
|
| 0.0683 | 1.0 | 110 | 0.0668 | |
|
| 0.0484 | 1.09 | 120 | 0.0707 | |
|
| 0.0567 | 1.18 | 130 | 0.0647 | |
|
| 0.0553 | 1.27 | 140 | 0.0635 | |
|
| 0.0515 | 1.36 | 150 | 0.0649 | |
|
| 0.0581 | 1.45 | 160 | 0.0615 | |
|
| 0.0486 | 1.54 | 170 | 0.0688 | |
|
| 0.052 | 1.63 | 180 | 0.0634 | |
|
| 0.0482 | 1.72 | 190 | 0.0638 | |
|
| 0.055 | 1.81 | 200 | 0.0606 | |
|
| 0.0469 | 1.9 | 210 | 0.0629 | |
|
| 0.0439 | 1.99 | 220 | 0.0682 | |
|
| 0.0278 | 2.08 | 230 | 0.0611 | |
|
| 0.0241 | 2.18 | 240 | 0.0702 | |
|
| 0.022 | 2.27 | 250 | 0.0783 | |
|
| 0.0203 | 2.36 | 260 | 0.0805 | |
|
| 0.0226 | 2.45 | 270 | 0.0801 | |
|
| 0.0207 | 2.54 | 280 | 0.0819 | |
|
| 0.0198 | 2.63 | 290 | 0.0807 | |
|
| 0.025 | 2.72 | 300 | 0.0785 | |
|
| 0.0286 | 2.81 | 310 | 0.0769 | |
|
| 0.0239 | 2.9 | 320 | 0.0764 | |
|
| 0.025 | 2.99 | 330 | 0.0764 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
