---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: PHI30512HMAB21H
  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. -->

# PHI30512HMAB21H

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: 0.1632

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 4.8935        | 0.09  | 10   | 1.6999          |
| 0.8206        | 0.18  | 20   | 0.2933          |
| 0.2869        | 0.27  | 30   | 0.2462          |
| 0.2573        | 0.36  | 40   | 0.2379          |
| 0.2401        | 0.45  | 50   | 0.2326          |
| 0.2293        | 0.54  | 60   | 0.2251          |
| 0.217         | 0.63  | 70   | 0.2020          |
| 0.2313        | 0.73  | 80   | 0.1992          |
| 0.2392        | 0.82  | 90   | 0.2193          |
| 0.214         | 0.91  | 100  | 0.1836          |
| 0.1548        | 1.0   | 110  | 0.1129          |
| 1.8394        | 1.09  | 120  | 0.7554          |
| 0.4491        | 1.18  | 130  | 0.1368          |
| 0.1653        | 1.27  | 140  | 0.0859          |
| 0.097         | 1.36  | 150  | 0.0882          |
| 1.1937        | 1.45  | 160  | 0.1699          |
| 0.2352        | 1.54  | 170  | 0.1636          |
| 0.1651        | 1.63  | 180  | 0.1664          |
| 0.1645        | 1.72  | 190  | 0.1658          |
| 0.1639        | 1.81  | 200  | 0.1645          |
| 0.1679        | 1.9   | 210  | 0.1646          |
| 0.1641        | 1.99  | 220  | 0.1642          |
| 0.1643        | 2.08  | 230  | 0.1634          |
| 0.1604        | 2.18  | 240  | 0.1629          |
| 0.16          | 2.27  | 250  | 0.1634          |
| 0.1631        | 2.36  | 260  | 0.1642          |
| 0.1617        | 2.45  | 270  | 0.1636          |
| 0.1617        | 2.54  | 280  | 0.1640          |
| 0.1619        | 2.63  | 290  | 0.1641          |
| 0.1632        | 2.72  | 300  | 0.1635          |
| 0.1634        | 2.81  | 310  | 0.1632          |
| 0.1617        | 2.9   | 320  | 0.1632          |
| 0.1661        | 2.99  | 330  | 0.1632          |


### Framework versions

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