Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on Countdown 8B Instruct (test)
Loading...
46.1
Accuracy
Top1 Prob + EOS Temp.
21.764
28.082
34.4
40.718
Apr 12, 2026
Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
Top1 Prob + EOS Temp.
Token Selection (T.S.)...
2026.04
46.1
Prob Margin + Both
Token Selection (T.S.)...
2026.04
45.3
Top1 Prob + Planner
Token Selection (T.S.)...
2026.04
44.1
Top1 Prob + Both
Token Selection (T.S.)...
2026.04
43.8
Temperature Sampling
Token Selection (T.S.)...
2026.04
43.4
Top1 Prob
Token Selection (T.S.)...
2026.04
42.2
Prob Margin
Token Selection (T.S.)...
2026.04
41.4
Init. Position
Token Selection (T.S.)...
2026.04
30.1
Ancestral Sampling
Token Selection (T.S.)...
2026.04
22.7
Feedback
Search any
task
Search any
task