---
license: mit
base_model: microsoft/phi-2
tags:
- generated_from_trainer
model-index:
- name: V0309P3
  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. -->

# V0309P3

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: 0.0857

## 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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 32
- 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: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.9399        | 0.09  | 10   | 0.3747          |
| 0.1877        | 0.17  | 20   | 0.0934          |
| 0.1061        | 0.26  | 30   | 0.0782          |
| 0.0988        | 0.34  | 40   | 0.0751          |
| 0.0879        | 0.43  | 50   | 0.0729          |
| 0.0823        | 0.51  | 60   | 0.0776          |
| 0.0735        | 0.6   | 70   | 0.0698          |
| 0.0775        | 0.68  | 80   | 0.0778          |
| 0.0716        | 0.77  | 90   | 0.0703          |
| 0.0687        | 0.85  | 100  | 0.0701          |
| 0.0718        | 0.94  | 110  | 0.0686          |
| 0.0679        | 1.02  | 120  | 0.0699          |
| 0.0579        | 1.11  | 130  | 0.0769          |
| 0.0559        | 1.19  | 140  | 0.0664          |
| 0.0527        | 1.28  | 150  | 0.0621          |
| 0.05          | 1.37  | 160  | 0.0753          |
| 0.0526        | 1.45  | 170  | 0.0628          |
| 0.0499        | 1.54  | 180  | 0.0685          |
| 0.0487        | 1.62  | 190  | 0.0711          |
| 0.0514        | 1.71  | 200  | 0.0705          |
| 0.0572        | 1.79  | 210  | 0.0724          |
| 0.0487        | 1.88  | 220  | 0.0700          |
| 0.0485        | 1.96  | 230  | 0.0693          |
| 0.0405        | 2.05  | 240  | 0.0706          |
| 0.0338        | 2.13  | 250  | 0.0833          |
| 0.0319        | 2.22  | 260  | 0.0897          |
| 0.0277        | 2.3   | 270  | 0.0941          |
| 0.0351        | 2.39  | 280  | 0.0891          |
| 0.0333        | 2.47  | 290  | 0.0839          |
| 0.0352        | 2.56  | 300  | 0.0867          |
| 0.0357        | 2.65  | 310  | 0.0839          |
| 0.0304        | 2.73  | 320  | 0.0842          |
| 0.0308        | 2.82  | 330  | 0.0859          |
| 0.0291        | 2.9   | 340  | 0.0856          |
| 0.0335        | 2.99  | 350  | 0.0857          |


### Framework versions

- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1