|
--- |
|
license: cc-by-nc-4.0 |
|
datasets: |
|
- kyujinpy/KOR-gugugu-platypus-set |
|
language: |
|
- en |
|
- ko |
|
base_model: |
|
- yanolja/KoSOLAR-10.7B-v0.2 |
|
pipeline_tag: text-generation |
|
--- |
|
|
|
# KoSOLAR-v0.2-gugutypus-10.7B |
|
|
|
<img src="logo.png" height=350, width=350> |
|
|
|
|
|
--- |
|
|
|
|
|
## Model Details |
|
|
|
**Model Developers** |
|
- DongGeon Lee ([oneonlee](https://huggingface.co/oneonlee)) |
|
|
|
**Model Architecture** |
|
- **KoSOLAR-v0.2-gugutypus-10.7B** is a instruction fine-tuned auto-regressive language model, based on the [SOLAR](https://huggingface.co/upstage/SOLAR-10.7B-v1.0) transformer architecture. |
|
|
|
**Base Model** |
|
- [yanolja/KoSOLAR-10.7B-v0.2](https://huggingface.co/yanolja/KoSOLAR-10.7B-v0.2) |
|
|
|
**Training Dataset** |
|
- [kyujinpy/KOR-gugugu-platypus-set](https://huggingface.co/datasets/kyujinpy/KOR-gugugu-platypus-set) |
|
|
|
**Environments** |
|
- Google Colab (Pro) |
|
- GPU : NVIDIA A100 40GB |
|
|
|
|
|
--- |
|
|
|
|
|
## Model comparisons |
|
|
|
- **Ko-LLM leaderboard (YYYY/MM/DD)** [[link]](https://huggingface.co/spaces/upstage/open-ko-llm-leaderboard) |
|
|
|
| Model | Average | Ko-ARC | Ko-HellaSwag | Ko-MMLU | Ko-TruthfulQA | Ko-CommonGen V2 | |
|
| --------------------- | ------- | ------ | ------------ | ------- | ------------- | --------------- | |
|
| **KoSOLAR-gugutypus** | NaN | NaN | NaN | NaN | NaN | NaN | |
|
|
|
<br> |
|
|
|
- **AI-Harness evaluation** [[link]](https://github.com/Beomi/ko-lm-evaluation-harness) |
|
|
|
|
|
| Model | Copa | Copa | HellaSwag | HellaSwag | BoolQ | BoolQ | Sentineg | Sentineg | |
|
| --------------------- | ------ | ------ | --------- | --------- | ------ | ------ | -------- | -------- | |
|
| | 0-shot | 5-shot | 0-shot | 5-shot | 0-shot | 5-shot | 0-shot | 5-shot | |
|
| **KoSOLAR-gugutypus** | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | |
|
|
|
|
|
--- |
|
|
|
|
|
## Implementation Code |
|
|
|
```python |
|
### KoSOLAR-gugutypus |
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
import torch |
|
|
|
repo = "oneonlee/KoSOLAR-v0.2-gugutypus-10.7B" |
|
model = AutoModelForCausalLM.from_pretrained( |
|
repo, |
|
return_dict=True, |
|
torch_dtype=torch.float16, |
|
device_map='auto' |
|
) |
|
tokenizer = AutoTokenizer.from_pretrained(repo) |
|
``` |