Mistral7B-PairRM-SPPO-Iter2 Benchmark Results

Mistral7B-PairRM-SPPO-Iter2 — benchmark results

Provider: Mistral. Released 2024-05-04. Access: Open.

Unified ELO 1363 ± 18, rank #1368 of 1776 rated models, from 15 benchmark results.

Benchmark	Score	Metric	Percentile
Open Chinese LLM - TruthfulQA MC	60.54	Accuracy (%)	84
Open Korean LLM Leaderboard	327.86	Average Score (%)	68.5
Open LLM Leaderboard - IFEval	44.46	Score	48.5
Open LLM Leaderboard - MuSR	9.8	Score	48.2
Open LLM Leaderboard - GPQA	5.15	Score	43.2
Open LLM Leaderboard - BBH	22.48	Score	33.7
Open Chinese LLM - ARC Challenge	47.7	Accuracy (%)	30.4
Open Chinese LLM - HellaSwag	56.25	Accuracy (%)	29.1
Open LLM Leaderboard - MMLU-Pro	18.63	Score	28.3
Open Chinese LLM - WinoGrande	61.17	Accuracy (%)	23.4
Open Chinese LLM Leaderboard	49.68	Average Score (%)	19.6
Open Chinese LLM - GSM8K	26.54	Accuracy (%)	19.3