---
license: mit
base_model: microsoft/phi-2
tags:
- generated_from_trainer
model-index:
- name: V0503HMA16H
  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. -->

# V0503HMA16H

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: 0.0683

## 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: 100
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.9481        | 0.09  | 10   | 0.6327          |
| 0.277         | 0.18  | 20   | 0.1138          |
| 0.1172        | 0.27  | 30   | 0.0926          |
| 0.0976        | 0.36  | 40   | 0.0808          |
| 0.0822        | 0.45  | 50   | 0.0722          |
| 0.0858        | 0.54  | 60   | 0.0703          |
| 0.0762        | 0.63  | 70   | 0.0685          |
| 0.0743        | 0.73  | 80   | 0.0798          |
| 0.0868        | 0.82  | 90   | 0.0676          |
| 0.0894        | 0.91  | 100  | 0.0750          |
| 0.0916        | 1.0   | 110  | 0.0740          |
| 0.0675        | 1.09  | 120  | 0.0908          |
| 0.0772        | 1.18  | 130  | 0.0846          |
| 0.0702        | 1.27  | 140  | 0.0765          |
| 0.0689        | 1.36  | 150  | 0.0718          |
| 0.0718        | 1.45  | 160  | 0.0746          |
| 0.0668        | 1.54  | 170  | 0.0629          |
| 0.0696        | 1.63  | 180  | 0.0693          |
| 0.0698        | 1.72  | 190  | 0.0690          |
| 0.0665        | 1.81  | 200  | 0.0678          |
| 0.0589        | 1.9   | 210  | 0.0709          |
| 0.0595        | 1.99  | 220  | 0.0708          |
| 0.0393        | 2.08  | 230  | 0.0743          |
| 0.0384        | 2.18  | 240  | 0.0757          |
| 0.0369        | 2.27  | 250  | 0.0737          |
| 0.0362        | 2.36  | 260  | 0.0765          |
| 0.0409        | 2.45  | 270  | 0.0756          |
| 0.0366        | 2.54  | 280  | 0.0714          |
| 0.0344        | 2.63  | 290  | 0.0714          |
| 0.0381        | 2.72  | 300  | 0.0688          |
| 0.0358        | 2.81  | 310  | 0.0687          |
| 0.035         | 2.9   | 320  | 0.0683          |
| 0.0361        | 2.99  | 330  | 0.0683          |


### Framework versions

- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.18.0
- Tokenizers 0.14.1