Share your thoughts, 1 month free Claude Pro on usSee more

Commonsense Reasoning on Winogrande (Accuracy, Speedup)

0.8624Accuracy

Exp-k=2 (7.5, 2.5)

Updated 1mo ago

Evaluation Results

Method	Links
Exp-k=2 (7.5, 2.5) 2025.12		0.8624	1.23
Exp-k=2 (7.5, 0) 2025.12		0.8616	1.54
Cosine (7.5, 0) 2025.12		0.8596	1.32
Linear (7.5, 0) 2025.12		0.8596	1.32
Cosine (7.5, 2.5) 2025.12		0.8591	1.23
Linear (7.5, 2.5) 2025.12		0.8585	1.22
Baseline 2025.12		0.858	1
Exp-k=16 (7.5, 0) 2025.12		0.8455	5
Exp-k=16 (7.5, 2.5) 2025.12		0.8449	2.29
Exp-k=8 (7.5, 0) 2025.12		0.7869	3.99
Baseline 2025.12		0.7798	1
Baseline 2025.05		0.737	-
Tensor RMS + C 2025.05		0.728	-
Block Absmax 2025.05		0.727	-
Prophet 2025.12		0.7258	1.03
Tensor RMS + Sp 2025.05		0.716	-
Channel Absmax 2025.05		0.693	-
Tensor Absmax 2025.05		0.665	-
Tensor RMS 2025.05		0.514	-