Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Commonsense Reasoning on HellaSwag (NLL)
Loading...
0.58
NLL
Base + TS
0.5528
0.7364
0.92
1.1036
May 26, 2026
NLL
Updated 1mo ago
Evaluation Results
Method
Method
Links
NLL
Base + TS
Backbone=QWEN2-7B, Tem...
2026.05
0.58
PPT (Distribution)
Backbone=QWEN2-7B, Tem...
2026.05
0.6
PPT (Distribution) + TS
Backbone=QWEN2-7B, Tem...
2026.05
0.6
PPT (Mean)
Backbone=QWEN2-7B, Tem...
2026.05
0.75
Base
Backbone=QWEN2-7B, Tem...
2026.05
0.84
PPT (Distribution) + TS
Backbone=LLAMA3-8B, Te...
2026.05
0.87
PPT (Distribution)
Backbone=LLAMA3-8B, Te...
2026.05
0.88
Base + TS
Backbone=LLAMA3-8B, Te...
2026.05
0.95
PPT (Mean)
Backbone=LLAMA3-8B, Te...
2026.05
0.98
Base
Backbone=LLAMA3-8B, Te...
2026.05
1.26
Feedback
Search any
task
Search any
task