---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: Phi0503HMA4
  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. -->

# Phi0503HMA4

This model is a fine-tuned version of [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0958

## 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: 80
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 4.3203        | 0.09  | 10   | 0.8287          |
| 0.439         | 0.18  | 20   | 0.2584          |
| 0.2853        | 0.27  | 30   | 0.2324          |
| 0.3519        | 0.36  | 40   | 0.2819          |
| 0.2642        | 0.45  | 50   | 1.2134          |
| 3.7539        | 0.54  | 60   | 4.0368          |
| 2.718         | 0.63  | 70   | 1.9987          |
| 1.3059        | 0.73  | 80   | 0.5453          |
| 0.534         | 0.82  | 90   | 0.4644          |
| 0.3535        | 0.91  | 100  | 0.2428          |
| 0.2618        | 1.0   | 110  | 0.2259          |
| 0.2222        | 1.09  | 120  | 0.2072          |
| 0.1939        | 1.18  | 130  | 0.1850          |
| 0.2268        | 1.27  | 140  | 0.1876          |
| 0.1966        | 1.36  | 150  | 0.1927          |
| 0.1942        | 1.45  | 160  | 0.1838          |
| 0.1937        | 1.54  | 170  | 0.1900          |
| 0.1808        | 1.63  | 180  | 0.1781          |
| 0.1785        | 1.72  | 190  | 0.1795          |
| 0.1737        | 1.81  | 200  | 0.1694          |
| 0.176         | 1.9   | 210  | 0.1708          |
| 0.1648        | 1.99  | 220  | 0.1751          |
| 0.1702        | 2.08  | 230  | 0.1576          |
| 0.1576        | 2.18  | 240  | 0.1524          |
| 0.1477        | 2.27  | 250  | 0.1389          |
| 0.1333        | 2.36  | 260  | 0.1172          |
| 0.1267        | 2.45  | 270  | 0.1126          |
| 0.1128        | 2.54  | 280  | 0.1016          |
| 0.1112        | 2.63  | 290  | 0.0995          |
| 0.1032        | 2.72  | 300  | 0.0976          |
| 0.1001        | 2.81  | 310  | 0.0971          |
| 0.0996        | 2.9   | 320  | 0.0957          |
| 0.0968        | 2.99  | 330  | 0.0958          |


### Framework versions

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