Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Commonsense Question Answering on ECQA (test)
Loading...
79.7
Accuracy
Llama2-70B
17.612
33.731
49.85
65.969
Apr 4, 2024
Accuracy
CT Unfaithfulness
CCT Faithfulness
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
CT Unfaithfulness
CCT Faithfulness
Llama2-70B
Prompt Order=predict-t...
2024.04
79.7
24.1
0.083
Llama2-70B
Prompt Order=explain-t...
2024.04
77.8
28.8
0.038
Llama2-13B
Prompt Order=explain-t...
2024.04
71.4
30.2
0.036
Llama2-13B
Prompt Order=predict-t...
2024.04
68
28.6
0.055
Llama2-7B
Prompt Order=explain-t...
2024.04
55.2
31.7
0.065
Llama2-7B
Prompt Order=predict-t...
2024.04
54.1
30.4
0.047
Random
2024.04
20
-
0
Feedback
Search any
task
Search any
task