---
license: mit
base_model: microsoft/phi-2
tags:
- generated_from_trainer
model-index:
- name: V0424HMA10
  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. -->

# V0424HMA10

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: 0.1353

## 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: 100
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.9203        | 0.09  | 10   | 0.5860          |
| 0.216         | 0.18  | 20   | 0.1276          |
| 0.1187        | 0.27  | 30   | 0.1089          |
| 0.1066        | 0.36  | 40   | 0.0864          |
| 0.0822        | 0.45  | 50   | 0.0775          |
| 0.0891        | 0.54  | 60   | 0.0866          |
| 0.0867        | 0.63  | 70   | 0.0769          |
| 0.0772        | 0.73  | 80   | 0.0991          |
| 0.0862        | 0.82  | 90   | 0.1365          |
| 4.6622        | 0.91  | 100  | 3.8668          |
| 1.4048        | 1.0   | 110  | 0.7169          |
| 0.5278        | 1.09  | 120  | 0.3863          |
| 0.3475        | 1.18  | 130  | 0.3058          |
| 0.2901        | 1.27  | 140  | 0.2546          |
| 0.2383        | 1.36  | 150  | 0.2151          |
| 0.1965        | 1.45  | 160  | 0.1826          |
| 0.1841        | 1.54  | 170  | 0.1697          |
| 0.1713        | 1.63  | 180  | 0.1678          |
| 0.1713        | 1.72  | 190  | 0.2457          |
| 0.1698        | 1.81  | 200  | 0.1620          |
| 0.1594        | 1.9   | 210  | 0.1489          |
| 0.1532        | 1.99  | 220  | 0.1470          |
| 0.1478        | 2.08  | 230  | 0.1530          |
| 0.1418        | 2.18  | 240  | 0.1405          |
| 0.1398        | 2.27  | 250  | 0.1367          |
| 0.1399        | 2.36  | 260  | 0.1384          |
| 0.1343        | 2.45  | 270  | 0.1368          |
| 0.1352        | 2.54  | 280  | 0.1354          |
| 0.1321        | 2.63  | 290  | 0.1372          |
| 0.1342        | 2.72  | 300  | 0.1354          |
| 0.1407        | 2.81  | 310  | 0.1351          |
| 0.1344        | 2.9   | 320  | 0.1352          |
| 0.1328        | 2.99  | 330  | 0.1353          |


### Framework versions

- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.18.0
- Tokenizers 0.14.1