Llama 3 8B ProLong 512k Base Benchmark Results

Llama 3 8B ProLong 512k Base — benchmark results

Provider: Meta. Released 2024-08-22. Access: Open.

Unified ELO 1372 ± 18, rank #1336 of 1776 rated models, from 15 benchmark results.

Benchmark	Score	Metric	Percentile
Open Korean LLM Leaderboard	391.71	Average Score (%)	76.1
Open LLM Leaderboard - MuSR	12.68	Score	69.5
Open LLM Leaderboard - IFEval	53.22	Score	63.3
Open LLM Leaderboard - BBH	29.85	Score	51.1
Open LLM Leaderboard - MMLU-Pro	25.88	Score	46.9
Open Chinese LLM - CMMLU	52.86	Accuracy (%)	44.5
Open Chinese LLM - WinoGrande	62.19	Accuracy (%)	37.8
Open LLM Leaderboard - MATH Level 5	6.87	Score	36.9
Open Chinese LLM - C-Eval Semantic	62.86	Accuracy (%)	32
Open Chinese LLM - GSM8K	34.19	Accuracy (%)	28.5
Open Chinese LLM - HellaSwag	56.14	Accuracy (%)	28.5
Open Chinese LLM - TruthfulQA MC	51.27	Accuracy (%)	26.7