Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Commonsense Reasoning on Hellaswag (fewshot=10, test)
Loading...
88.57
Accuracy
PhiMoE (FP16 baseline)
87.582
87.8385
88.095
88.3515
Apr 20, 2026
Accuracy
Average Token Drop Rate (%)
Average Padded Tokens (%)
Accuracy Degradation (%)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Average Token Drop Rate (%)
Average Padded Tokens (%)
Accuracy Degradation (%)
PhiMoE (FP16 baseline)
Precision=FP16, Shot c...
2026.04
88.57
-
-
-
NPUMoE
Precision=FP16, Shot c...
2026.04
87.62
10.63
34.77
-1.07
Feedback
Search any
task
Search any
task