List your company

Helpful Dialogue on Anthropic HH-RLHF helpful core250 (test)

18.93Reward ScoreTEA
1.363365.9239310.484515.04507May 11, 2026

Evaluation results

MethodLinks
TEAN=1282026.0518.931.8171.33667.22.430.4
GRPON=2562026.0518.354—————
TEAN=2562026.0518.3541.81.26464.83.631.6
TEAN=642026.0517.6741.941.50272.8225.2
GRPON=1282026.0517.113—————
TEAN=322026.0516.2241.981.60575.22.422.4
GRPON=642026.0515.734—————
TEAN=162026.0514.5551.9111.58377.6220.4
GRPON=322026.0514.244—————
GRPON=162026.0512.644—————
TEAN=82026.0512.551.7921.48278.4219.6
GRPON=82026.0510.758—————
TEAN=42026.0510.121.6341.32178220
GRPON=42026.058.485—————
TEAN=22026.057.0811.3751.07176.8221.2
GRPON=22026.055.707—————
TEAN=12026.053.0170.9780.68371.6226.4
GRPON=12026.052.039—————