NeuralSynthesis-7B-v0.4-slerp Benchmark Results

NeuralSynthesis-7B-v0.4-slerp — benchmark results

Provider: Other. Released 2024-04-12. Access: Open.

Unified ELO 1338 ± 31, rank #1445 of 1776 rated models, from 20 benchmark results.

Benchmark	Score	Metric	Percentile
Open Chinese LLM - TruthfulQA MC	63.44	Accuracy (%)	93.5
Open LLM Leaderboard - MuSR	13.06	Score	71.8
Open Chinese LLM - ARC Challenge	54.01	Accuracy (%)	61.9
Open LLM Leaderboard - BBH	32	Score	59.6
Open CoT - LSAT Analytical Reasoning	4.35	CoT Gain (%)	56.9
Open Chinese LLM - HellaSwag	60.69	Accuracy (%)	56.7
Open Chinese LLM - GSM8K	46.55	Accuracy (%)	53.7
Open Chinese LLM Leaderboard	56.69	Average Score (%)	51.3
Open Chinese LLM - WinoGrande	62.98	Accuracy (%)	51.2
Open LLM Leaderboard - IFEval	39.47	Score	39.3
Open LLM Leaderboard - MMLU-Pro	22.7	Score	37.9
Open LLM Leaderboard - MATH Level 5	6.27	Score	34.3