BBEH
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
BBEH
78.8Accuracy
76
BBEH
97DCS AUROC
35
BBEH
3.78KL Divergence (nats)
35
BBEH (test)
-0.99Delta IB (nats)
35
BBEH
72.6Same-task Accuracy
35
BBEH
58.9Accuracy
27
BBEH
15.7pass@1
23
BBEH
55.2Accuracy (Causal Reasoning)
14
BBEH (test)
34.5Accuracy
14
BBEH (val)
66.4Top-1 Acc
14
BBEH
66.4Top-1 Accuracy
14
BBEH mini
14.8Pass@1
13
BBEH
9.94Pass@1
12
BBEH
81.2Accuracy
12
BBEH Mini
17.8Accuracy
11
BBEH
75.8Accuracy
7
BBEH
67.2AOC
7
BBEH
0.665AOC
7
BBEH mini
17Accuracy
6
bbeh
52.37Score
5
BBEH Web of Lies
90.12Accuracy
3
BBEH Dyck Languages
91.25Accuracy
3
BBEH Disambiguation QA
65.41Accuracy
3