---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: Phi0503HMA15
  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. -->

# Phi0503HMA15

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: 0.0616

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 4.4844        | 0.09  | 10   | 0.8736          |
| 0.3931        | 0.18  | 20   | 0.2417          |
| 0.2471        | 0.27  | 30   | 0.2360          |
| 0.2276        | 0.36  | 40   | 0.2169          |
| 0.2082        | 0.45  | 50   | 0.1854          |
| 0.2197        | 0.54  | 60   | 0.2227          |
| 0.2251        | 0.63  | 70   | 0.2017          |
| 0.2114        | 0.73  | 80   | 0.2120          |
| 0.2076        | 0.82  | 90   | 0.1874          |
| 0.1868        | 0.91  | 100  | 0.1686          |
| 0.1725        | 1.0   | 110  | 0.1657          |
| 0.163         | 1.09  | 120  | 0.1645          |
| 0.164         | 1.18  | 130  | 0.1647          |
| 0.1737        | 1.27  | 140  | 0.1626          |
| 0.1665        | 1.36  | 150  | 0.1629          |
| 0.1662        | 1.45  | 160  | 0.1651          |
| 0.1425        | 1.54  | 170  | 0.0893          |
| 0.1374        | 1.63  | 180  | 0.0857          |
| 0.129         | 1.72  | 190  | 0.1095          |
| 0.0855        | 1.81  | 200  | 0.0848          |
| 0.0678        | 1.9   | 210  | 0.0823          |
| 0.0723        | 1.99  | 220  | 0.0818          |
| 0.062         | 2.08  | 230  | 0.0671          |
| 0.0583        | 2.18  | 240  | 0.0690          |
| 0.0553        | 2.27  | 250  | 0.0685          |
| 0.0511        | 2.36  | 260  | 0.0654          |
| 0.0528        | 2.45  | 270  | 0.0659          |
| 0.0493        | 2.54  | 280  | 0.0657          |
| 0.0479        | 2.63  | 290  | 0.0650          |
| 0.0483        | 2.72  | 300  | 0.0642          |
| 0.0598        | 2.81  | 310  | 0.0628          |
| 0.0528        | 2.9   | 320  | 0.0615          |
| 0.0529        | 2.99  | 330  | 0.0616          |


### Framework versions

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