---
license: mit
base_model: microsoft/phi-2
tags:
- generated_from_trainer
model-index:
- name: V0309P1
  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. -->

# V0309P1

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: 0.0820

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 1.4262        | 0.09  | 10   | 0.1204          |
| 0.1236        | 0.17  | 20   | 0.0907          |
| 0.1031        | 0.26  | 30   | 0.0766          |
| 0.0896        | 0.34  | 40   | 0.0691          |
| 0.0871        | 0.43  | 50   | 0.0719          |
| 0.0821        | 0.51  | 60   | 0.0751          |
| 0.0749        | 0.6   | 70   | 0.0676          |
| 0.0809        | 0.68  | 80   | 0.0624          |
| 0.068         | 0.77  | 90   | 0.0591          |
| 0.062         | 0.85  | 100  | 0.0666          |
| 0.0712        | 0.94  | 110  | 0.0643          |
| 0.0679        | 1.02  | 120  | 0.0600          |
| 0.0488        | 1.11  | 130  | 0.0758          |
| 0.0498        | 1.19  | 140  | 0.0573          |
| 0.0451        | 1.28  | 150  | 0.0649          |
| 0.0434        | 1.37  | 160  | 0.0692          |
| 0.0449        | 1.45  | 170  | 0.0639          |
| 0.0401        | 1.54  | 180  | 0.0697          |
| 0.0477        | 1.62  | 190  | 0.0633          |
| 0.0492        | 1.71  | 200  | 0.0609          |
| 0.0489        | 1.79  | 210  | 0.0632          |
| 0.0422        | 1.88  | 220  | 0.0679          |
| 0.0417        | 1.96  | 230  | 0.0633          |
| 0.034         | 2.05  | 240  | 0.0678          |
| 0.0247        | 2.13  | 250  | 0.0700          |
| 0.0234        | 2.22  | 260  | 0.0766          |
| 0.0187        | 2.3   | 270  | 0.0816          |
| 0.0231        | 2.39  | 280  | 0.0841          |
| 0.0245        | 2.47  | 290  | 0.0859          |
| 0.024         | 2.56  | 300  | 0.0848          |
| 0.0253        | 2.65  | 310  | 0.0847          |
| 0.0202        | 2.73  | 320  | 0.0841          |
| 0.0242        | 2.82  | 330  | 0.0814          |
| 0.0187        | 2.9   | 340  | 0.0820          |
| 0.0217        | 2.99  | 350  | 0.0820          |


### Framework versions

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