KyujinHan/Poly-platypus-ko

Polyglot-Ko를 기반으로 KOpen-platypus dataset을 이용하여 fine-tuning한 model

stars

commits

Oct 3, 2023

updated

README

Poly-platypus-ko

KO-platypus Korean-Open-platypus를 활용하여 Polyglot-ko를 fine-tuning한 Korean-Platypus model

본 연구는 (주)마커와 (주)미디어그룹사람과숲의 오픈소스 LLM 연구 컨소시엄에서 진행되었습니다.

Model description

Poly-platypus-ko-12.8b🥮:
KOpen-Platypus🥮:

Polyglot-ko + KO-Platypus

CoTy-platypus-ko-12.8b:
KoCoT_2000:

Polyglot-ko + CoT-llama2-ko

KO-platypus2 & CoT-llama2-ko

KO-platypus2
CoT-llama2-ko

Performance

When I evaluated Ko-Platy, I used this repo.
And, implement below code.

# In colab,
!python main.py \
    --model gpt2 \ 
    --model_args pretrained=..your_model_name.. \
    --tasks kobest_hellaswag,kobest_copa,kobest_boolq,kobest_sentineg \
    --device cuda:0 \
    --num_fewshot 0 # 5, 10, 25, ...

COPA (F1)

Model	0-shot	5-shot	10-shot	50-shot
Polyglot-ko-5.8b	0.7745	0.7676	0.7775	0.7887
Polyglot-ko-12.8b	0.7937	0.8108	0.8037	0.8369
Llama-2-Ko-7b 20B	0.7388	0.7626	0.7808	0.7979
Llama-2-Ko-7b 40B	0.7436	0.7927	0.8037	0.8259
KO-platypus2-7B-EX	0.7509	0.7899	0.8029	0.8290
KoT-platypus2-7B	0.7517	0.7868	0.8009	0.8239
Poly-platypus-ko-12.8b(ours)	0.7876	0.8099	0.8008	0.8239

HellaSwag (F1)

Model	0-shot	5-shot	10-shot	50-shot
Polyglot-ko-5.8b	0.5976	0.5998	0.5979	0.6208
Polyglot-ko-12.8b	0.5954	0.6306	0.6098	0.6118
Llama-2-Ko-7b 20B	0.4518	0.4668	0.4726	0.4828
Llama-2-Ko-7b 40B	0.4562	0.4657	0.4698	0.4774
KO-platypus2-7B-EX	0.4571	0.4461	0.4371	0.4525
KoT-platypus2-7B	0.4432	0.4382	0.4550	0.4534
Poly-platypus-ko-12.8b(ours)	0.4838	0.4858	0.5005	0.5062

BoolQ (F1)

Model	0-shot	5-shot	10-shot	50-shot
Polyglot-ko-5.8b	0.4356	0.5698	0.5187	0.5236
Polyglot-ko-12.8b	0.4818	0.6041	0.6289	0.6448
Llama-2-Ko-7b 20B	0.3607	0.6797	0.6801	0.6622
Llama-2-Ko-7b 40B	0.5786	0.6977	0.7084	0.7144
KO-platypus2-7B-EX	0.6028	0.6979	0.7016	0.6988
KoT-platypus2-7B	0.6142	0.6757	0.6839	0.6878
Poly-platypus-ko-12.8b(ours)	0.4888	0.6520	0.6568	0.6835

SentiNeg (F1)

Model	0-shot	5-shot	10-shot	50-shot
Polyglot-ko-5.8b	0.3394	0.8841	0.8808	0.9521
Polyglot-ko-12.8b	0.9117	0.9015	0.9345	0.9723
Llama-2-Ko-7b 20B	0.4855	0.8295	0.8711	0.8513
Llama-2-Ko-7b 40B	0.4594	0.7611	0.7276	0.9370
KO-platypus2-7B-EX	0.5821	0.7653	0.7991	0.8643
KoT-platypus2-7B	0.6127	0.7199	0.7531	0.8381
Poly-platypus-ko-12.8b(ours)	0.8490	0.9597	0.9723	0.9847

References

Kopen-Platypus🥮
MarkrAI/kyujin-Poly-platypus-ko-12.8b🥮
Platypus
Polyglot-ko-12.8b
ko-lm-evaluation-harness

Additional info about image

I made the image, inspired by Platypus-LLM.
I used Playground AI, then applying prompt engineering. (For example, img2img, guidance etc...)

When I made Poly-platypus image, I use prompt like above image.

Contributors

KyujinHan

8 commits

Similar repos

KyujinHan/KoT-platypus

CoT 방식을 활용하여 KO-platypus fine-tuning

same org

Marker-Inc-Korea/KO-Platypus

[KO-Platy🥮] Korean-Open-platypus를 활용하여 llama-2-ko를 fine-tuning한 KO-platypus model

MarkrAI/kyujin-Poly-platypus-ko-12.8b

Model

kyujinpy/KoT-platypus2-7B

Model

MarkrAI/kyujin-CoTy-platypus-ko-12.8b

Model

Marker-Inc-Korea/Korean-OpenOrca

OpenOrca-KO dataset을 활용하여 llama2를 fine-tuning한 Korean-OpenOrca

Marker-Inc-Korea/CoT-llama2

Chain-of-thought 방식을 활용하여 llama2를 fine-tuning

kyujinpy/CoT-llama-2k-7b

Model

KyujinHan/Poly-platypus-ko

Polyglot-Ko를 기반으로 KOpen-platypus dataset을 이용하여 fine-tuning한 model

stars

commits

Oct 3, 2023

updated

README

Poly-platypus-ko

KO-platypus Korean-Open-platypus를 활용하여 Polyglot-ko를 fine-tuning한 Korean-Platypus model

본 연구는 (주)마커와 (주)미디어그룹사람과숲의 오픈소스 LLM 연구 컨소시엄에서 진행되었습니다.

Model description

Poly-platypus-ko-12.8b🥮:
KOpen-Platypus🥮:

Polyglot-ko + KO-Platypus

CoTy-platypus-ko-12.8b:
KoCoT_2000:

Polyglot-ko + CoT-llama2-ko

KO-platypus2 & CoT-llama2-ko

KO-platypus2
CoT-llama2-ko

Performance

When I evaluated Ko-Platy, I used this repo.
And, implement below code.

# In colab,
!python main.py \
    --model gpt2 \ 
    --model_args pretrained=..your_model_name.. \
    --tasks kobest_hellaswag,kobest_copa,kobest_boolq,kobest_sentineg \
    --device cuda:0 \
    --num_fewshot 0 # 5, 10, 25, ...

COPA (F1)

Model	0-shot	5-shot	10-shot	50-shot
Polyglot-ko-5.8b	0.7745	0.7676	0.7775	0.7887
Polyglot-ko-12.8b	0.7937	0.8108	0.8037	0.8369
Llama-2-Ko-7b 20B	0.7388	0.7626	0.7808	0.7979
Llama-2-Ko-7b 40B	0.7436	0.7927	0.8037	0.8259
KO-platypus2-7B-EX	0.7509	0.7899	0.8029	0.8290
KoT-platypus2-7B	0.7517	0.7868	0.8009	0.8239
Poly-platypus-ko-12.8b(ours)	0.7876	0.8099	0.8008	0.8239

HellaSwag (F1)

Model	0-shot	5-shot	10-shot	50-shot
Polyglot-ko-5.8b	0.5976	0.5998	0.5979	0.6208
Polyglot-ko-12.8b	0.5954	0.6306	0.6098	0.6118
Llama-2-Ko-7b 20B	0.4518	0.4668	0.4726	0.4828
Llama-2-Ko-7b 40B	0.4562	0.4657	0.4698	0.4774
KO-platypus2-7B-EX	0.4571	0.4461	0.4371	0.4525
KoT-platypus2-7B	0.4432	0.4382	0.4550	0.4534
Poly-platypus-ko-12.8b(ours)	0.4838	0.4858	0.5005	0.5062

BoolQ (F1)

Model	0-shot	5-shot	10-shot	50-shot
Polyglot-ko-5.8b	0.4356	0.5698	0.5187	0.5236
Polyglot-ko-12.8b	0.4818	0.6041	0.6289	0.6448
Llama-2-Ko-7b 20B	0.3607	0.6797	0.6801	0.6622
Llama-2-Ko-7b 40B	0.5786	0.6977	0.7084	0.7144
KO-platypus2-7B-EX	0.6028	0.6979	0.7016	0.6988
KoT-platypus2-7B	0.6142	0.6757	0.6839	0.6878
Poly-platypus-ko-12.8b(ours)	0.4888	0.6520	0.6568	0.6835

SentiNeg (F1)

Model	0-shot	5-shot	10-shot	50-shot
Polyglot-ko-5.8b	0.3394	0.8841	0.8808	0.9521
Polyglot-ko-12.8b	0.9117	0.9015	0.9345	0.9723
Llama-2-Ko-7b 20B	0.4855	0.8295	0.8711	0.8513
Llama-2-Ko-7b 40B	0.4594	0.7611	0.7276	0.9370
KO-platypus2-7B-EX	0.5821	0.7653	0.7991	0.8643
KoT-platypus2-7B	0.6127	0.7199	0.7531	0.8381
Poly-platypus-ko-12.8b(ours)	0.8490	0.9597	0.9723	0.9847