Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Logical Reasoning on Logic (test)
Loading...
0.722
AUROC
Post-solution probe
0.62424
0.64962
0.675
0.70038
Jul 8, 2026
AUROC
ECE
Recovery Rate
Updated 16d ago
Evaluation Results
Method
Method
Links
AUROC
ECE
Recovery Rate
Post-solution probe
Probe Type=Post-solution
2026.07
0.722
0.17
-
Distilled future confidence predictor
Probe Type=Distilled
2026.07
0.696
0.184
-
Pre-solution probe
Probe Type=Pre-solution
2026.07
0.628
0.202
-
Feedback
Search any
task
Search any
task