Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Knowledge Evaluation on WikiText (eval)
Loading...
0.777
BPB
Disagreement
0.77116
0.81058
0.85
0.88942
Jan 31, 2026
BPB
Updated 4d ago
Evaluation Results
Method
Method
Links
BPB
Disagreement
Protocol=Training-time
2026.01
0.777
CAD
Protocol=Inference-time
2026.01
0.779
Baseline
Architecture=LoRA, Los...
2026.01
0.784
GAME-LoRA
Protocol=Training-time
2026.01
0.786
ME
Protocol=Training-time
2026.01
0.825
ActDec
Protocol=Inference-time
2026.01
0.923
Feedback
Search any
task
Search any
task