---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: PHI30512HMAB16H
  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. -->

# PHI30512HMAB16H

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: 0.1633

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 4.7915        | 0.09  | 10   | 1.7498          |
| 0.9756        | 0.18  | 20   | 0.2822          |
| 0.2692        | 0.27  | 30   | 0.2313          |
| 0.2426        | 0.36  | 40   | 0.2245          |
| 0.2219        | 0.45  | 50   | 0.2097          |
| 0.2025        | 0.54  | 60   | 0.2170          |
| 0.2107        | 0.63  | 70   | 0.2090          |
| 0.2164        | 0.73  | 80   | 0.2058          |
| 0.1826        | 0.82  | 90   | 0.1733          |
| 0.1676        | 0.91  | 100  | 0.1635          |
| 0.2174        | 1.0   | 110  | 0.1705          |
| 0.165         | 1.09  | 120  | 0.1654          |
| 0.1654        | 1.18  | 130  | 0.1675          |
| 0.1758        | 1.27  | 140  | 0.1648          |
| 0.1669        | 1.36  | 150  | 0.1644          |
| 0.161         | 1.45  | 160  | 0.1656          |
| 0.1644        | 1.54  | 170  | 0.1633          |
| 0.1635        | 1.63  | 180  | 0.1627          |
| 0.1641        | 1.72  | 190  | 0.1680          |
| 0.1642        | 1.81  | 200  | 0.1642          |
| 0.1656        | 1.9   | 210  | 0.1636          |
| 0.1625        | 1.99  | 220  | 0.1648          |
| 0.1635        | 2.08  | 230  | 0.1633          |
| 0.1606        | 2.18  | 240  | 0.1627          |
| 0.1597        | 2.27  | 250  | 0.1631          |
| 0.1625        | 2.36  | 260  | 0.1636          |
| 0.161         | 2.45  | 270  | 0.1635          |
| 0.1613        | 2.54  | 280  | 0.1638          |
| 0.1615        | 2.63  | 290  | 0.1639          |
| 0.1628        | 2.72  | 300  | 0.1636          |
| 0.163         | 2.81  | 310  | 0.1634          |
| 0.1617        | 2.9   | 320  | 0.1633          |
| 0.166         | 2.99  | 330  | 0.1633          |


### Framework versions

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