---
license: cc-by-4.0
---
# **KoQuality-Polyglot-5.8b**


KoQuality-Polyglot-5.8b is a fine-tuned version of [EleutherAI/polyglot-ko-5.8b](https://huggingface.co/EleutherAI/polyglot-ko-5.8b) on [KoQuality dataset](https://huggingface.co/datasets/DILAB-HYU/KoQuality), which is curated by proposed method (len_group=5, k=100, n=0.01, method=ppl_sampling). 


## Overall Average accuracy score of the KoBEST datasets

We use [KoBEST benchmark](https://huggingface.co/datasets/skt/kobest_v1) datasets(BoolQ, COPA, HellaSwag, SentiNeg, WiC) to compare the performance of our best model and other models accuracy. Our model outperforms other models in the average accuracy score of the KoBEST datasets.
<img src=https://cdn-uploads.huggingface.co/production/uploads/650fecfd247f564485f8fbcf/t5x4PphoNb-tW3iCzXXHT.png style="max-width: 500px; width: 300%"/>




| Model | 0-shot | 1-shot | 2-shot | 5-shot | 10-shot
| --- | --- | --- | --- | --- | --- |
| polyglot-ko-5.8b | 0.4734 | 0.5929 | 0.6120 | 0.6388 | 0.6295
| koalpcaca-polyglot-5.8b | 0.4731 | 0.5284 | 0.5721 | 0.6054 | 0.6042
| kullm-polyglot-5.8b | 0.4415 | 0.6030 | 0.5849 | 0.6252 | 0.6451
| koquality-polyglot-5.8b | 0.4530 | 0.6050 | 0.6351 | 0.6420 | 0.6457

## Evaluation results
### COPA (F1)
<img src=https://cdn-uploads.huggingface.co/production/uploads/650fecfd247f564485f8fbcf/QAie0x99S8-KEKvK0I_uZ.png style="max-width: 400px; width: 200%"/>

### BoolQ (F1)
<img src=https://cdn-uploads.huggingface.co/production/uploads/650fecfd247f564485f8fbcf/CtEWEQ5BBS05V9cDWA7kp.png style="max-width: 400px; width: 200%"/>

### HellaSwag (F1)
<img src=https://cdn-uploads.huggingface.co/production/uploads/650fecfd247f564485f8fbcf/cHws6qWkDlTfs5GVcQvtN.png style="max-width: 400px; width: 200%"/>


### SentiNeg (F1)
<img src=https://cdn-uploads.huggingface.co/production/uploads/650fecfd247f564485f8fbcf/VEG15XXOIbzJyQAusLa4B.png style="max-width: 400px; width: 200%"/>


### WiC (F1)
<img src=https://cdn-uploads.huggingface.co/production/uploads/650fecfd247f564485f8fbcf/hV-uADJiydkVQOyYysej9.png style="max-width: 400px; width: 200%"/>



## Training hyperparameters
- learning_rate: 5e-5
- train_batch_size: 4
- seed: 42
- distributed_type: multi-GPU (A100 80G)
- num_devices: 4
- gradient_accumulation_steps: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2.0

## Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu117
- Datasets 2.11.0
- deepspeed 0.9.5

## Citation

```
@misc{2023koqaulity,
  title = {KoQuality: Curation of High-quality Instruction Data for Korean Language Models},
  author = {Na, Yohan and Kim, Dahye and Chae, Dong-Kyu},
  journal={Proceedings of the 35th Annual Conference on Human and Cognitive Language Technology (HCLT 2023)},
  pages={},
  year = {2023},
}
```


<br>