Pr_Llama2.7B-Sh5K_Wi5K_Ne5K_Ct5K-Lr05_Ep4 Benchmark Results

Pr_Llama2.7B-Sh5K_Wi5K_Ne5K_Ct5K-Lr05_Ep4: benchmark results

Provider: Other. Access: Open.

Unified ELO 1375 ± 17, rank #2087 of 2656 rated models, from 10 benchmark results.

Benchmark	Score	Metric	Percentile
Open Korean LLM - KorNAT-Harmless	64.11	Normalized accuracy (%)	70.7
Open Korean LLM - Ko-GPQA-Diamond	24.24	Normalized accuracy (%)	62.4
Open Korean LLM - Ko-EQ-Bench	2.38	EQ-Bench score	48.6
Open Korean LLM - Ko-IFEval	26.7	Strict accuracy, prompt/instruction mean (%)	40
Open Korean LLM - KorNAT-Helpful	45.62	Normalized accuracy (%)	38.2
Open Korean LLM - KorNAT-CKA	21.21	Normalized accuracy (%)	33.5
Open Korean LLM Leaderboard	31.87	Average Score (%)	29.8
Open Ko-LLM Leaderboard	31.87	Average (%)	29.6
Open Korean LLM - Ko-GSM8K (flexible extract)	6.67	Exact match, flexible extract (%)	28.5
Open Korean LLM - Ko-Winogrande	51.38	Accuracy (%)	6.4