---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: PHI30515HMA1H
  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. -->

# PHI30515HMA1H

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: 0.0747

## 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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 32
- 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 7.2832        | 0.09  | 10   | 2.7337          |
| 1.7648        | 0.18  | 20   | 0.3745          |
| 0.3839        | 0.27  | 30   | 0.2589          |
| 0.3285        | 0.36  | 40   | 0.2520          |
| 0.3202        | 0.45  | 50   | 0.2229          |
| 0.6502        | 0.54  | 60   | 0.2693          |
| 0.3048        | 0.63  | 70   | 0.1647          |
| 0.2068        | 0.73  | 80   | 0.1318          |
| 0.1411        | 0.82  | 90   | 0.1621          |
| 0.1775        | 0.91  | 100  | 0.0975          |
| 0.1835        | 1.0   | 110  | 0.0954          |
| 0.1014        | 1.09  | 120  | 0.0876          |
| 0.1148        | 1.18  | 130  | 0.0976          |
| 0.1506        | 1.27  | 140  | 0.0760          |
| 0.128         | 1.36  | 150  | 0.0750          |
| 0.0883        | 1.45  | 160  | 0.0736          |
| 0.0913        | 1.54  | 170  | 0.0692          |
| 0.0795        | 1.63  | 180  | 0.0681          |
| 0.0927        | 1.72  | 190  | 0.0669          |
| 0.087         | 1.81  | 200  | 0.0667          |
| 0.0606        | 1.9   | 210  | 0.0682          |
| 0.0627        | 1.99  | 220  | 0.0679          |
| 0.0441        | 2.08  | 230  | 0.0705          |
| 0.0543        | 2.18  | 240  | 0.0813          |
| 0.0413        | 2.27  | 250  | 0.0839          |
| 0.0414        | 2.36  | 260  | 0.0775          |
| 0.0462        | 2.45  | 270  | 0.0756          |
| 0.0411        | 2.54  | 280  | 0.0763          |
| 0.0392        | 2.63  | 290  | 0.0768          |
| 0.0407        | 2.72  | 300  | 0.0771          |
| 0.0508        | 2.81  | 310  | 0.0755          |
| 0.0577        | 2.9   | 320  | 0.0746          |
| 0.0431        | 2.99  | 330  | 0.0747          |


### Framework versions

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