---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: Phi0503HMA3
  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. -->

# Phi0503HMA3

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: 0.0755

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 4.2281        | 0.09  | 10   | 0.6893          |
| 0.3554        | 0.18  | 20   | 0.2337          |
| 0.2494        | 0.27  | 30   | 0.2261          |
| 0.2206        | 0.36  | 40   | 0.1916          |
| 0.213         | 0.45  | 50   | 0.1778          |
| 0.1546        | 0.54  | 60   | 0.1014          |
| 0.1079        | 0.63  | 70   | 0.0987          |
| 0.0823        | 0.73  | 80   | 0.0974          |
| 0.0902        | 0.82  | 90   | 0.0855          |
| 0.0772        | 0.91  | 100  | 0.0706          |
| 0.076         | 1.0   | 110  | 0.0844          |
| 0.0666        | 1.09  | 120  | 0.0719          |
| 0.0634        | 1.18  | 130  | 0.0803          |
| 0.0711        | 1.27  | 140  | 0.0697          |
| 0.0638        | 1.36  | 150  | 0.0679          |
| 0.0665        | 1.45  | 160  | 0.0687          |
| 0.0635        | 1.54  | 170  | 0.0664          |
| 0.0605        | 1.63  | 180  | 0.0674          |
| 0.0554        | 1.72  | 190  | 0.0641          |
| 0.0604        | 1.81  | 200  | 0.0623          |
| 0.0567        | 1.9   | 210  | 0.0664          |
| 0.0528        | 1.99  | 220  | 0.0693          |
| 0.0327        | 2.08  | 230  | 0.0751          |
| 0.0273        | 2.18  | 240  | 0.0921          |
| 0.0225        | 2.27  | 250  | 0.0998          |
| 0.0254        | 2.36  | 260  | 0.0898          |
| 0.0331        | 2.45  | 270  | 0.0737          |
| 0.021         | 2.54  | 280  | 0.0749          |
| 0.0256        | 2.63  | 290  | 0.0767          |
| 0.0274        | 2.72  | 300  | 0.0765          |
| 0.0299        | 2.81  | 310  | 0.0760          |
| 0.0242        | 2.9   | 320  | 0.0754          |
| 0.0273        | 2.99  | 330  | 0.0755          |


### Framework versions

- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1