Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Feature Matching on GPT2 Layer 0 match with Layer 11
Loading...
1.39
LLM Eval Score
Semantic Optimal Transport
1.13
1.1975
1.265
1.3325
May 27, 2026
LLM Eval Score
Accuracy
Cross-Entropy (CE)
Vector Error (VE)
Updated 6d ago
Evaluation Results
Method
Method
Links
LLM Eval Score
Accuracy
Cross-Entropy (CE)
Vector Error (VE)
Semantic Optimal Transport
2026.05
1.39
17.2
3.828
0.9111
FeatFlow
2026.05
1.34
12.4
3.855
0.9095
Match
2026.05
1.27
10
3.833
0.911
Attrib
2026.05
1.18
6.2
3.837
0.9105
FeatFlow ML
2026.05
1.16
5.4
3.831
0.911
Match ML
2026.05
1.14
6
3.911
0.9063
Feedback
Search any
task
Search any
task