Llama 3 8B Magpie Align V0.3 Benchmark Results

Llama 3 8B Magpie Align V0.3 — benchmark results

Provider: Meta. Released 2024-07-15. Access: Open.

Unified ELO 1360 ± 16, rank #1382 of 1776 rated models, from 15 benchmark results.

Benchmark	Score	Metric	Percentile
Open Korean LLM Leaderboard	345.71	Average Score (%)	70.3
Open Chinese LLM - TruthfulQA MC	54.4	Accuracy (%)	56.7
Open Chinese LLM - C-Eval Semantic	69.92	Accuracy (%)	51.3
Open LLM Leaderboard - IFEval	44.97	Score	49.4
Open Chinese LLM - HellaSwag	60.06	Accuracy (%)	49.3
Open Chinese LLM - ARC Challenge	51.11	Accuracy (%)	45
Open Chinese LLM - CMMLU	52.64	Accuracy (%)	42.7
Open LLM Leaderboard - MMLU-Pro	23.71	Score	41.4
Open Chinese LLM - WinoGrande	62.19	Accuracy (%)	37.8
Open Chinese LLM Leaderboard	55.06	Average Score (%)	37.7
Open LLM Leaderboard - BBH	24.31	Score	37.1
Open LLM Leaderboard - MATH Level 5	5.66	Score	31.3