|
--- |
|
license: mit |
|
base_model: microsoft/phi-2 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: V0410MPTEST2 |
|
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. --> |
|
|
|
# V0410MPTEST2 |
|
|
|
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: 0.0710 |
|
|
|
## 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: 20 |
|
- num_epochs: 2 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 0.2909 | 0.18 | 20 | 0.1525 | |
|
| 0.135 | 0.36 | 40 | 0.1083 | |
|
| 0.1067 | 0.54 | 60 | 0.0895 | |
|
| 0.084 | 0.73 | 80 | 0.0829 | |
|
| 0.0802 | 0.91 | 100 | 0.0767 | |
|
| 0.0785 | 1.09 | 120 | 0.0748 | |
|
| 0.0759 | 1.27 | 140 | 0.0744 | |
|
| 0.0797 | 1.45 | 160 | 0.0723 | |
|
| 0.07 | 1.63 | 180 | 0.0714 | |
|
| 0.0742 | 1.81 | 200 | 0.0712 | |
|
| 0.0704 | 1.99 | 220 | 0.0710 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
