---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: Phi0503HMA19
  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. -->

# Phi0503HMA19

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: 0.0782

## 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: 60
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 3.8941        | 0.09  | 10   | 0.5252          |
| 0.304         | 0.18  | 20   | 0.2303          |
| 0.286         | 0.27  | 30   | 0.2215          |
| 0.2077        | 0.36  | 40   | 0.1718          |
| 0.153         | 0.45  | 50   | 0.1491          |
| 0.1408        | 0.54  | 60   | 0.1229          |
| 0.1063        | 0.63  | 70   | 0.1014          |
| 0.1079        | 0.73  | 80   | 0.0884          |
| 0.0837        | 0.82  | 90   | 0.0777          |
| 0.081         | 0.91  | 100  | 0.0710          |
| 0.0749        | 1.0   | 110  | 0.0747          |
| 0.0557        | 1.09  | 120  | 0.0717          |
| 0.0566        | 1.18  | 130  | 0.0735          |
| 0.0588        | 1.27  | 140  | 0.0713          |
| 0.0535        | 1.36  | 150  | 0.0778          |
| 0.0672        | 1.45  | 160  | 0.0677          |
| 0.0548        | 1.54  | 170  | 0.0702          |
| 0.0536        | 1.63  | 180  | 0.0633          |
| 0.0474        | 1.72  | 190  | 0.0653          |
| 0.0537        | 1.81  | 200  | 0.0633          |
| 0.0438        | 1.9   | 210  | 0.0669          |
| 0.0483        | 1.99  | 220  | 0.0669          |
| 0.0233        | 2.08  | 230  | 0.0749          |
| 0.0245        | 2.18  | 240  | 0.0840          |
| 0.017         | 2.27  | 250  | 0.0868          |
| 0.0166        | 2.36  | 260  | 0.0866          |
| 0.0258        | 2.45  | 270  | 0.0823          |
| 0.015         | 2.54  | 280  | 0.0805          |
| 0.018         | 2.63  | 290  | 0.0811          |
| 0.0227        | 2.72  | 300  | 0.0795          |
| 0.0217        | 2.81  | 310  | 0.0788          |
| 0.0196        | 2.9   | 320  | 0.0784          |
| 0.0198        | 2.99  | 330  | 0.0782          |


### Framework versions

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