---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: Phi0503HMA18
  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. -->

# Phi0503HMA18

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: 0.0856

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 3.766         | 0.09  | 10   | 0.4613          |
| 0.3024        | 0.18  | 20   | 0.2538          |
| 0.2469        | 0.27  | 30   | 0.2178          |
| 0.261         | 0.36  | 40   | 0.2338          |
| 0.482         | 0.45  | 50   | 0.2783          |
| 0.2062        | 0.54  | 60   | 0.1451          |
| 0.1438        | 0.63  | 70   | 0.1608          |
| 0.1457        | 0.73  | 80   | 0.1216          |
| 0.103         | 0.82  | 90   | 0.1031          |
| 0.1028        | 0.91  | 100  | 0.0730          |
| 0.0779        | 1.0   | 110  | 0.0722          |
| 0.0572        | 1.09  | 120  | 0.0660          |
| 0.0543        | 1.18  | 130  | 0.0833          |
| 0.0669        | 1.27  | 140  | 0.0697          |
| 0.0573        | 1.36  | 150  | 0.0711          |
| 0.0636        | 1.45  | 160  | 0.0659          |
| 0.0579        | 1.54  | 170  | 0.0680          |
| 0.0595        | 1.63  | 180  | 0.0651          |
| 0.0541        | 1.72  | 190  | 0.0692          |
| 0.0579        | 1.81  | 200  | 0.0626          |
| 0.0487        | 1.9   | 210  | 0.0663          |
| 0.0501        | 1.99  | 220  | 0.0655          |
| 0.0269        | 2.08  | 230  | 0.0733          |
| 0.0241        | 2.18  | 240  | 0.0912          |
| 0.0172        | 2.27  | 250  | 0.1040          |
| 0.0171        | 2.36  | 260  | 0.0972          |
| 0.0242        | 2.45  | 270  | 0.0892          |
| 0.0153        | 2.54  | 280  | 0.0907          |
| 0.0193        | 2.63  | 290  | 0.0904          |
| 0.0252        | 2.72  | 300  | 0.0876          |
| 0.0245        | 2.81  | 310  | 0.0862          |
| 0.0202        | 2.9   | 320  | 0.0857          |
| 0.0207        | 2.99  | 330  | 0.0856          |


### Framework versions

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