---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: PHI30512HMAB6H
  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. -->

# PHI30512HMAB6H

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: 0.0698

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 4.4764        | 0.09  | 10   | 1.7039          |
| 0.6874        | 0.18  | 20   | 0.4076          |
| 0.3403        | 0.27  | 30   | 0.2637          |
| 0.2668        | 0.36  | 40   | 0.2667          |
| 0.8232        | 0.45  | 50   | 3.1025          |
| 0.8435        | 0.54  | 60   | 0.2217          |
| 0.2786        | 0.63  | 70   | 0.2215          |
| 0.2253        | 0.73  | 80   | 0.2019          |
| 0.1871        | 0.82  | 90   | 0.1830          |
| 0.1871        | 0.91  | 100  | 0.1695          |
| 0.2185        | 1.0   | 110  | 0.2040          |
| 0.1712        | 1.09  | 120  | 0.1659          |
| 0.4398        | 1.18  | 130  | 0.2223          |
| 1.8534        | 1.27  | 140  | 2.5467          |
| 1.547         | 1.36  | 150  | 0.7915          |
| 0.6568        | 1.45  | 160  | 0.4273          |
| 0.3954        | 1.54  | 170  | 0.4106          |
| 0.3571        | 1.63  | 180  | 0.3610          |
| 0.2652        | 1.72  | 190  | 0.1875          |
| 0.207         | 1.81  | 200  | 0.1718          |
| 0.1613        | 1.9   | 210  | 0.1431          |
| 0.1411        | 1.99  | 220  | 0.1365          |
| 0.1256        | 2.08  | 230  | 0.1349          |
| 0.1252        | 2.18  | 240  | 0.1040          |
| 0.1131        | 2.27  | 250  | 0.0991          |
| 0.1023        | 2.36  | 260  | 0.0855          |
| 0.0836        | 2.45  | 270  | 0.0725          |
| 0.082         | 2.54  | 280  | 0.0696          |
| 0.0906        | 2.63  | 290  | 0.0696          |
| 0.075         | 2.72  | 300  | 0.0694          |
| 0.0766        | 2.81  | 310  | 0.0683          |
| 0.0741        | 2.9   | 320  | 0.0700          |
| 0.0664        | 2.99  | 330  | 0.0698          |


### Framework versions

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