---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: PHI30512HMAB26H
  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. -->

# PHI30512HMAB26H

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: 0.0713

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 4.8606        | 0.09  | 10   | 1.2529          |
| 0.634         | 0.18  | 20   | 0.2571          |
| 0.2918        | 0.27  | 30   | 0.2613          |
| 0.26          | 0.36  | 40   | 0.2799          |
| 0.2618        | 0.45  | 50   | 0.2381          |
| 0.2224        | 0.54  | 60   | 0.2114          |
| 0.2238        | 0.63  | 70   | 0.2278          |
| 0.2084        | 0.73  | 80   | 0.1813          |
| 0.1347        | 0.82  | 90   | 0.1057          |
| 0.0992        | 0.91  | 100  | 0.1193          |
| 0.0989        | 1.0   | 110  | 0.0853          |
| 0.0869        | 1.09  | 120  | 0.0838          |
| 0.0762        | 1.18  | 130  | 0.0755          |
| 0.0791        | 1.27  | 140  | 0.0743          |
| 0.0777        | 1.36  | 150  | 0.0775          |
| 0.0797        | 1.45  | 160  | 0.0712          |
| 0.0729        | 1.54  | 170  | 0.0685          |
| 0.0705        | 1.63  | 180  | 0.0706          |
| 0.0672        | 1.72  | 190  | 0.0751          |
| 0.0734        | 1.81  | 200  | 0.0688          |
| 0.0646        | 1.9   | 210  | 0.0709          |
| 0.0596        | 1.99  | 220  | 0.0763          |
| 0.0421        | 2.08  | 230  | 0.0864          |
| 0.0425        | 2.18  | 240  | 0.0845          |
| 0.0358        | 2.27  | 250  | 0.0775          |
| 0.0343        | 2.36  | 260  | 0.0763          |
| 0.0449        | 2.45  | 270  | 0.0717          |
| 0.0323        | 2.54  | 280  | 0.0723          |
| 0.0319        | 2.63  | 290  | 0.0724          |
| 0.0369        | 2.72  | 300  | 0.0727          |
| 0.0402        | 2.81  | 310  | 0.0717          |
| 0.0358        | 2.9   | 320  | 0.0713          |
| 0.0369        | 2.99  | 330  | 0.0713          |


### Framework versions

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