Llama-3-Base-8B-SFT-RRHF Benchmark Results

Llama-3-Base-8B-SFT-RRHF — benchmark results

Provider: Meta. Released 2024-07-06. Access: Open.

Unified ELO 1360 ± 32, rank #1383 of 1776 rated models, from 7 benchmark results.

Benchmark	Score	Metric	Percentile
Open Korean LLM Leaderboard	291.13	Average Score (%)	66.3
Open LLM Leaderboard - GPQA	7.38	Score	62.2
Open LLM Leaderboard - BBH	23.66	Score	35.8
Open LLM Leaderboard - MuSR	7.56	Score	35.2
Open LLM Leaderboard - MMLU-Pro	20.99	Score	33.1
Open LLM Leaderboard - IFEval	33.57	Score	31.7
Open LLM Leaderboard - MATH Level 5	4.53	Score	25.7