---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: Phi0503HMA12
  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. -->

# Phi0503HMA12

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: 0.1480

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 4.2962        | 0.09  | 10   | 0.8255          |
| 0.3877        | 0.18  | 20   | 0.2353          |
| 0.264         | 0.27  | 30   | 0.2552          |
| 0.236         | 0.36  | 40   | 0.2439          |
| 0.2233        | 0.45  | 50   | 0.2179          |
| 0.211         | 0.54  | 60   | 0.2097          |
| 0.1896        | 0.63  | 70   | 0.1825          |
| 1.1817        | 0.73  | 80   | 3.3232          |
| 4.4778        | 0.82  | 90   | 2.4034          |
| 1.5484        | 0.91  | 100  | 0.7261          |
| 0.5632        | 1.0   | 110  | 0.4332          |
| 0.3744        | 1.09  | 120  | 0.5888          |
| 0.3942        | 1.18  | 130  | 0.3256          |
| 0.3126        | 1.27  | 140  | 0.2506          |
| 0.2203        | 1.36  | 150  | 0.2069          |
| 0.2076        | 1.45  | 160  | 0.1817          |
| 0.1926        | 1.54  | 170  | 0.1959          |
| 0.1984        | 1.63  | 180  | 0.1913          |
| 0.1895        | 1.72  | 190  | 0.1800          |
| 0.1813        | 1.81  | 200  | 0.1858          |
| 0.184         | 1.9   | 210  | 0.1763          |
| 0.1716        | 1.99  | 220  | 0.1732          |
| 0.1728        | 2.08  | 230  | 0.1689          |
| 0.1671        | 2.18  | 240  | 0.1650          |
| 0.1618        | 2.27  | 250  | 0.1610          |
| 0.1577        | 2.36  | 260  | 0.1565          |
| 0.1506        | 2.45  | 270  | 0.1517          |
| 0.153         | 2.54  | 280  | 0.1512          |
| 0.1499        | 2.63  | 290  | 0.1512          |
| 0.1485        | 2.72  | 300  | 0.1484          |
| 0.1563        | 2.81  | 310  | 0.1477          |
| 0.1499        | 2.9   | 320  | 0.1480          |
| 0.148         | 2.99  | 330  | 0.1480          |


### Framework versions

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