---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: PHI30512HMAB20H
  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. -->

# PHI30512HMAB20H

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: 0.0752

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 4.2937        | 0.09  | 10   | 0.8823          |
| 0.4617        | 0.18  | 20   | 0.2708          |
| 0.2681        | 0.27  | 30   | 2.7303          |
| 0.9935        | 0.36  | 40   | 0.2454          |
| 0.2512        | 0.45  | 50   | 0.2272          |
| 0.2279        | 0.54  | 60   | 0.2115          |
| 0.2067        | 0.63  | 70   | 0.2056          |
| 0.2419        | 0.73  | 80   | 0.1810          |
| 0.1545        | 0.82  | 90   | 0.0988          |
| 0.0955        | 0.91  | 100  | 0.0863          |
| 0.0846        | 1.0   | 110  | 0.0745          |
| 0.073         | 1.09  | 120  | 0.0728          |
| 0.0688        | 1.18  | 130  | 0.0799          |
| 0.0731        | 1.27  | 140  | 0.0723          |
| 0.0702        | 1.36  | 150  | 0.0740          |
| 0.0793        | 1.45  | 160  | 0.0680          |
| 0.0662        | 1.54  | 170  | 0.0651          |
| 0.0666        | 1.63  | 180  | 0.0636          |
| 0.0605        | 1.72  | 190  | 0.0640          |
| 0.0678        | 1.81  | 200  | 0.0666          |
| 0.0568        | 1.9   | 210  | 0.0702          |
| 0.0568        | 1.99  | 220  | 0.0660          |
| 0.0351        | 2.08  | 230  | 0.0769          |
| 0.032         | 2.18  | 240  | 0.0946          |
| 0.0288        | 2.27  | 250  | 0.0879          |
| 0.0276        | 2.36  | 260  | 0.0766          |
| 0.0316        | 2.45  | 270  | 0.0777          |
| 0.0269        | 2.54  | 280  | 0.0781          |
| 0.0265        | 2.63  | 290  | 0.0789          |
| 0.0322        | 2.72  | 300  | 0.0770          |
| 0.0362        | 2.81  | 310  | 0.0756          |
| 0.0294        | 2.9   | 320  | 0.0749          |
| 0.0277        | 2.99  | 330  | 0.0752          |


### Framework versions

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