Llama 3 8B SFR Iterative DPO R Benchmark Results

Llama 3 8B SFR Iterative DPO R — benchmark results

Provider: Meta. Released 2024-05-09. Access: Open.

Unified ELO 1369 ± 9, rank #1349 of 1776 rated models, from 16 benchmark results.

Benchmark	Score	Metric	Percentile
Open CoT - LSAT Analytical Reasoning	5.22	CoT Gain (%)	68.3
Open CoT - LogiQA 2	7.7	CoT Gain (%)	55.7
Open CoT Leaderboard	7.43	Average CoT Gain (%)	51.9
Open CoT - LSAT Reading Comprehension	11.52	CoT Gain (%)	49.6
Open LLM Leaderboard - BBH	29.15	Score	48.7
Open LLM Leaderboard - MATH Level 5	9.14	Score	44.7
Open CoT - LogiQA	3.67	CoT Gain (%)	43.1
Open LLM Leaderboard - MMLU-Pro	24.14	Score	42.9
Open LLM Leaderboard - GPQA	5.03	Score	42.5
Open CoT - LSAT Logical Reasoning	9.02	CoT Gain (%)	41.2
Open LLM Leaderboard - IFEval	38.16	Score	37.5
Open LLM Leaderboard - MuSR	5.55	Score	26.9