(주)미디어그룹사람과숲과 (주)마커의 LLM 연구 컨소시엄에서 개발된 모델입니다
The license is cc-by-nc-sa-4.0.
KOR-Orca-Platypus-kiwi🥝
Model Details
Model Developers Kyujin Han (kyujinpy)
Model Architecture
ko-platypus-kiwi-13B is an auto-regressive language model based on the LLaMA2 transformer architecture.
Base Model hyunseoki/ko-en-llama2-13b
Training Dataset
I used kyujinpy/KOR-Orca-Platypus-kiwi.
Model comparisons
| Model | Average | Ko-ARC | Ko-HellaSwag | Ko-MMLU | Ko-TruthfulQA | Ko-CommonGen V2 |
|---|---|---|---|---|---|---|
| ko-platypus-kiwi-13B🥝 | 48.97 | 42.41 | 54.29 | 41.98 | 40.05 | 66.12 |
Implementation Code
### KO-Platypus
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
repo = "kyujinpy/ko-platypus-kiwi-13B"
OpenOrca = AutoModelForCausalLM.from_pretrained(
repo,
return_dict=True,
torch_dtype=torch.float16,
device_map='auto'
)
OpenOrca_tokenizer = AutoTokenizer.from_pretrained(repo)