Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Instruction Tuning on Alpaca instruction-tuning 52k
Loading...
116
Pairwise Winning Score
GRADFILTERING
-3.496
27.527
58.55
89.573
Jan 20, 2026
Pairwise Winning Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Pairwise Winning Score
GRADFILTERING
Base Model=LLaMA2-7B,...
2026.01
116
GRADFILTERING
Base Model=LLaMA2-7B,...
2026.01
116
GRADFILTERING
Base Model=LLaMA2-7B,...
2026.01
115
GRADFILTERING
Base Model=LLaMA2-13B,...
2026.01
113
GRADFILTERING
Base Model=LLaMA2-7B,...
2026.01
112
Superfiltering
Base Model=LLaMA2-7B,...
2026.01
110
GRADFILTERING
Base Model=LLaMA2-13B,...
2026.01
110
GRADFILTERING
Base Model=LLaMA2-7B,...
2026.01
107
GRADFILTERING
Base Model=LLaMA2-7B,...
2026.01
106
GRADFILTERING
Base Model=LLaMA2-7B,...
2026.01
106
Superfiltering
Base Model=LLaMA2-7B,...
2026.01
102
Superfiltering
Base Model=LLaMA2-7B,...
2026.01
102
GRADFILTERING
Base Model=LLaMA2-7B,...
2026.01
101
Random Split
Base Model=LLaMA2-7B,...
2026.01
97
Random Split
Base Model=LLaMA2-7B,...
2026.01
93
Random Split
Base Model=LLaMA2-7B,...
2026.01
92
GRADFILTERING
Base Model=LLaMA2-13B,...
2026.01
1.2
GRADFILTERING
Base Model=LLaMA2-13B,...
2026.01
1.19
GRADFILTERING
Base Model=LLaMA2-13B,...
2026.01
1.1
Feedback
Search any
task
Search any
task