Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Reasoning on Winogrande 5-shot
Loading...
0.749
Normalized Log Accuracy
HATified-SFT
0.59612
0.63581
0.6755
0.71519
Mar 16, 2026
Normalized Log Accuracy
Compression Ratio
Updated 1mo ago
Evaluation Results
Method
Method
Links
Normalized Log Accuracy
Compression Ratio
HATified-SFT
shots=5-shot
2026.03
0.749
-
Llama-Instruct
shots=5-shot
2026.03
0.697
-
HATified
shot count=5-shot, DPO...
2026.03
0.687
5.16
Llama-3.1-8B-TFree-HAT-SFT
shots=5, SFT=true
2026.03
0.677
5.16
T-Free
shot count=5-shot, DPO...
2026.03
0.676
5.16
Llama
shot count=5-shot, DPO...
2026.03
0.673
4.91
Tülu
shot count=5-shot, DPO...
2026.03
0.635
4.91
Llama-3.1-Tulu-3-8B-SFT
shots=5, SFT=true
2026.03
0.602
4.91
Feedback
Search any
task
Search any
task