Einstein-v6.1-Llama3-8B Benchmark Results

Einstein-v6.1-Llama3-8B — benchmark results

Provider: Other. Released 2024-04-19. Access: Open.

Unified ELO 1397 ± 8, rank #1241 of 1776 rated models, from 14 benchmark results.

Benchmark	Score	Metric	Percentile
Open Chinese LLM - WinoGrande	65.04	Accuracy (%)	70.8
Open LLM Leaderboard - MuSR	11.23	Score	58.3
Open LLM Leaderboard - IFEval	45.68	Score	51
Open LLM Leaderboard - BBH	29.38	Score	49.5
Open Chinese LLM - TruthfulQA MC	53.51	Accuracy (%)	48.5
Open Chinese LLM - CMMLU	53.24	Accuracy (%)	48.4
Open Chinese LLM - GSM8K	42.15	Accuracy (%)	43
Open Chinese LLM - C-Eval Semantic	67.01	Accuracy (%)	41.7
Open LLM Leaderboard - MMLU-Pro	23.68	Score	41.3
Open Chinese LLM - HellaSwag	58.46	Accuracy (%)	39.8
Open Chinese LLM Leaderboard	55.26	Average Score (%)	38.9
Open LLM Leaderboard - GPQA	4.25	Score	36.8