---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: Phi0503HMA8
  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. -->

# Phi0503HMA8

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: 0.1630

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 4.6142        | 0.09  | 10   | 1.3143          |
| 0.5476        | 0.18  | 20   | 0.2468          |
| 0.3629        | 0.27  | 30   | 0.2346          |
| 0.2414        | 0.36  | 40   | 0.2269          |
| 0.2169        | 0.45  | 50   | 0.1917          |
| 0.1925        | 0.54  | 60   | 0.1844          |
| 0.1977        | 0.63  | 70   | 0.1833          |
| 0.1755        | 0.73  | 80   | 0.1650          |
| 0.1689        | 0.82  | 90   | 0.1627          |
| 0.1552        | 0.91  | 100  | 0.7844          |
| 0.2892        | 1.0   | 110  | 0.1149          |
| 1.1144        | 1.09  | 120  | 0.1357          |
| 0.1033        | 1.18  | 130  | 0.0829          |
| 0.098         | 1.27  | 140  | 0.0898          |
| 0.0863        | 1.36  | 150  | 0.0845          |
| 0.0913        | 1.45  | 160  | 0.0791          |
| 0.0782        | 1.54  | 170  | 0.0708          |
| 0.0804        | 1.63  | 180  | 0.0786          |
| 0.089         | 1.72  | 190  | 0.2288          |
| 0.3087        | 1.81  | 200  | 0.1845          |
| 0.449         | 1.9   | 210  | 0.3669          |
| 0.7395        | 1.99  | 220  | 0.3523          |
| 0.5132        | 2.08  | 230  | 0.1956          |
| 0.1939        | 2.18  | 240  | 0.1647          |
| 0.1612        | 2.27  | 250  | 0.1673          |
| 0.1638        | 2.36  | 260  | 0.1636          |
| 0.1617        | 2.45  | 270  | 0.1634          |
| 0.1617        | 2.54  | 280  | 0.1640          |
| 0.1626        | 2.63  | 290  | 0.1641          |
| 0.1635        | 2.72  | 300  | 0.1634          |
| 0.1638        | 2.81  | 310  | 0.1632          |
| 0.162         | 2.9   | 320  | 0.1630          |
| 0.1659        | 2.99  | 330  | 0.1630          |


### Framework versions

- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.0