---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: Phi0511B2
  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. -->

# Phi0511B2

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: 0.0718

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 5.4836        | 0.09  | 10   | 5.4324          |
| 5.4542        | 0.18  | 20   | 5.2447          |
| 4.9491        | 0.27  | 30   | 4.1820          |
| 3.3816        | 0.36  | 40   | 2.1650          |
| 1.3373        | 0.45  | 50   | 0.5823          |
| 0.4096        | 0.54  | 60   | 0.1966          |
| 0.1569        | 0.63  | 70   | 0.1428          |
| 0.1378        | 0.73  | 80   | 0.1266          |
| 0.1135        | 0.82  | 90   | 0.1250          |
| 0.127         | 0.91  | 100  | 0.1191          |
| 0.1078        | 1.0   | 110  | 0.1075          |
| 0.0958        | 1.09  | 120  | 0.0948          |
| 0.0877        | 1.18  | 130  | 0.0946          |
| 0.0995        | 1.27  | 140  | 0.0841          |
| 0.0822        | 1.36  | 150  | 0.1001          |
| 0.0901        | 1.45  | 160  | 0.0814          |
| 0.0754        | 1.54  | 170  | 0.0833          |
| 0.0806        | 1.63  | 180  | 0.1000          |
| 0.0763        | 1.72  | 190  | 0.0800          |
| 0.0787        | 1.81  | 200  | 0.0787          |
| 0.0637        | 1.9   | 210  | 0.0753          |
| 0.0657        | 1.99  | 220  | 0.0799          |
| 0.0605        | 2.08  | 230  | 0.0777          |
| 0.0625        | 2.18  | 240  | 0.0740          |
| 0.0544        | 2.27  | 250  | 0.0743          |
| 0.0572        | 2.36  | 260  | 0.0736          |
| 0.0593        | 2.45  | 270  | 0.0739          |
| 0.0575        | 2.54  | 280  | 0.0729          |
| 0.0584        | 2.63  | 290  | 0.0737          |
| 0.057         | 2.72  | 300  | 0.0728          |
| 0.0622        | 2.81  | 310  | 0.0720          |
| 0.0566        | 2.9   | 320  | 0.0718          |
| 0.0594        | 2.99  | 330  | 0.0718          |


### Framework versions

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