Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
OOD Detection on LLaMa 1 (test)
Loading...
0.924
AUROC
EigenTrack
0.69208
0.75229
0.8125
0.87271
Jan 24, 2026
AUROC
Updated 4d ago
Evaluation Results
Method
Method
Links
AUROC
EigenTrack
Model Scale=7B
2026.01
0.924
ODIN
Model Scale=7B
2026.01
0.921
Cosine Distance
Model Scale=7B
2026.01
0.92
EigenTrack
Model Scale=3B
2026.01
0.892
Energy Score
Model Scale=7B
2026.01
0.89
Cosine Distance
Model Scale=3B
2026.01
0.877
EigenTrack
Model Scale=1B
2026.01
0.855
Energy Score
Model Scale=3B
2026.01
0.852
ODIN
Model Scale=3B
2026.01
0.842
Energy Score
Model Scale=1B
2026.01
0.832
Cosine Distance
Model Scale=1B
2026.01
0.819
ODIN
Model Scale=1B
2026.01
0.801
Max Softmax Prob
Model Scale=7B
2026.01
0.72
Max Softmax Prob
Model Scale=3B
2026.01
0.71
Max Softmax Prob
Model Scale=1B
2026.01
0.701
Feedback
Search any
task
Search any
task