Edit model card

EEVE-Math-10.8B

EEVE-Math ν”„λ‘œμ νŠΈλŠ”

에 λŒ€ν•œ λ‚΄μš©μ„ ν¬κ΄„ν•˜κ³  μžˆμŠ΅λ‹ˆλ‹€.

이 λͺ¨λΈμ€ orca-math-word-problems-193k-korean 데이터셋을 μ΄μš©ν•˜μ—¬ ν•™μŠ΅λ˜μ—ˆμŠ΅λ‹ˆλ‹€. 응닡 쀑 μΌλΆ€λŠ” LaTeX ν˜•μ‹μ„ μ΄μš©ν•˜μ—¬ κ²°κ³Όλ₯Ό λ°˜ν™˜ν•˜μ§€λ§Œ, μ™„μ„±λœ ν˜•μ‹μ΄ 아닐 수 μžˆμŠ΅λ‹ˆλ‹€. ν˜„μž¬ M1 stageκΉŒμ§€ μ§„ν–‰λ˜μ—ˆμŠ΅λ‹ˆλ‹€.

Model gsm8k-ko(pass@1)
Base 0.4049
SFT(M1) 0.508
SFT(M1) -> SFT 0.539
SFT(M1) -> KTO(M2) -
SFT -> KTO(M2) -> KTO(final) -

Specifications

  • SFT(M1) -> SFT 단계

Base Model

yanolja/EEVE-Korean-10.8B-v1.0

Dataset

orca-math-word-problems-193k-korean

Evaluation

gsm8k-ko, kobest

git clone https://github.com/kuotient/lm-evaluation-harness.git
cd lm-evaluation-harness
pip install -e .
lm_eval --model hf \
    --model_args pretrained=yanolja/EEVE-Korean-Instruct-2.8B-v1.0 \
    --tasks gsm8k-ko \
    --device cuda:0 \
    --batch_size auto:4
Model gsm8k(pass@1) boolq(acc) copa(acc) hellaswag(acc) Overall
yanolja/EEVE-Korean-10.8B-v1.0 0.4049 - - - -
yanolja/EEVE-Korean-Instruct-10.8B-v1.0 0.4511 0.8668 0.7450 0.4940 0.6392
EEVE-Math-10.8B 0.5390 0.8027 0.7260 0.4760 0.6359
EEVE-Instruct-Math-10.8B 0.4845 0.8519 0.7410 0.4980 0.6439
Downloads last month
159
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for kuotient/EEVE-Math-10.8B

Finetuned
(9)
this model

Dataset used to train kuotient/EEVE-Math-10.8B

Evaluation results