---
license: mit
base_model: microsoft/phi-2
tags:
- generated_from_trainer
model-index:
- name: V0422MADP4C
  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. -->

# V0422MADP4C

This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0645

## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 5.2111        | 0.09  | 10   | 1.5167          |
| 2.2548        | 0.18  | 20   | 0.1595          |
| 0.4128        | 0.27  | 30   | 0.1403          |
| 0.169         | 0.36  | 40   | 0.1317          |
| 0.1338        | 0.45  | 50   | 0.1031          |
| 0.1151        | 0.54  | 60   | 0.0947          |
| 0.0959        | 0.63  | 70   | 0.0844          |
| 0.0888        | 0.73  | 80   | 0.0796          |
| 0.0899        | 0.82  | 90   | 0.0807          |
| 0.0912        | 0.91  | 100  | 0.0758          |
| 0.0926        | 1.0   | 110  | 0.0711          |
| 0.0795        | 1.09  | 120  | 0.0754          |
| 0.0992        | 1.18  | 130  | 0.0936          |
| 0.0917        | 1.27  | 140  | 0.0777          |
| 0.2413        | 1.36  | 150  | 0.1380          |
| 0.1347        | 1.45  | 160  | 0.0987          |
| 0.1056        | 1.54  | 170  | 0.0780          |
| 0.0903        | 1.63  | 180  | 0.0736          |
| 0.0827        | 1.72  | 190  | 0.0713          |
| 0.0864        | 1.81  | 200  | 0.0839          |
| 0.0796        | 1.9   | 210  | 0.0808          |
| 0.0782        | 1.99  | 220  | 0.0747          |
| 0.0716        | 2.08  | 230  | 0.0691          |
| 0.0689        | 2.18  | 240  | 0.0679          |
| 0.0669        | 2.27  | 250  | 0.0660          |
| 0.068         | 2.36  | 260  | 0.0649          |
| 0.0658        | 2.45  | 270  | 0.0655          |
| 0.0639        | 2.54  | 280  | 0.0654          |
| 0.0602        | 2.63  | 290  | 0.0647          |
| 0.0619        | 2.72  | 300  | 0.0647          |
| 0.0687        | 2.81  | 310  | 0.0648          |
| 0.0624        | 2.9   | 320  | 0.0645          |
| 0.0711        | 2.99  | 330  | 0.0645          |


### Framework versions

- Transformers 4.36.0.dev0
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.14.1