List your company

Acceptability Judgment on SLING Chinese (test)

67AUCProbe Score
54.5257.766164.24May 6, 2026

Evaluation results

MethodLinks
Probe ScoreModel=Gemma-2-9B, Zero-shot=true2026.056775
Probe ScoreModel=Gemma-2-2B, Zero-shot=true2026.056373
Probe ScoreModel=Llama-3.1-8B, Zero-shot=true2026.056069
LM LogprobModel=Gemma-2-9B, Zero-shot=true2026.055862
Probe ScoreModel=Llama-3.2-1B, Zero-shot=true2026.055763
LM LogprobModel=Llama-3.1-8B, Zero-shot=true2026.055765
LM LogprobModel=Gemma-2-2B, Zero-shot=true2026.055761
LM LogprobModel=Llama-3.2-1B, Zero-shot=true2026.055559