---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: PHI30511HMA10H
  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. -->

# PHI30511HMA10H

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: 0.0859

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 3.9089        | 0.09  | 10   | 1.2284          |
| 0.5288        | 0.18  | 20   | 0.1720          |
| 0.1533        | 0.27  | 30   | 0.1436          |
| 0.1373        | 0.36  | 40   | 0.1243          |
| 0.1281        | 0.45  | 50   | 0.1184          |
| 0.1123        | 0.54  | 60   | 0.0945          |
| 0.0973        | 0.63  | 70   | 0.1022          |
| 0.0916        | 0.73  | 80   | 0.0787          |
| 0.0665        | 0.82  | 90   | 0.0685          |
| 0.0746        | 0.91  | 100  | 0.0688          |
| 0.0656        | 1.0   | 110  | 0.0695          |
| 0.0472        | 1.09  | 120  | 0.0709          |
| 0.0456        | 1.18  | 130  | 0.0672          |
| 0.0554        | 1.27  | 140  | 0.0644          |
| 0.046         | 1.36  | 150  | 0.0653          |
| 0.0531        | 1.45  | 160  | 0.0609          |
| 0.0486        | 1.54  | 170  | 0.0649          |
| 0.0493        | 1.63  | 180  | 0.0616          |
| 0.0464        | 1.72  | 190  | 0.0636          |
| 0.0522        | 1.81  | 200  | 0.0612          |
| 0.0423        | 1.9   | 210  | 0.0606          |
| 0.0457        | 1.99  | 220  | 0.0606          |
| 0.0224        | 2.08  | 230  | 0.0676          |
| 0.022         | 2.18  | 240  | 0.0788          |
| 0.016         | 2.27  | 250  | 0.0873          |
| 0.0137        | 2.36  | 260  | 0.0910          |
| 0.0204        | 2.45  | 270  | 0.0903          |
| 0.0146        | 2.54  | 280  | 0.0899          |
| 0.0172        | 2.63  | 290  | 0.0890          |
| 0.0206        | 2.72  | 300  | 0.0870          |
| 0.02          | 2.81  | 310  | 0.0863          |
| 0.0186        | 2.9   | 320  | 0.0860          |
| 0.0175        | 2.99  | 330  | 0.0859          |


### Framework versions

- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.18.0
- Tokenizers 0.14.1