INDEX
Explanations
possessive pronouns and abstract nouns
New Auto-Interp
Negative Logits
An
0.47
Only
0.46
Option
0.44
Engineer
0.43
Request
0.43
F
0.43
Eine
0.42
Unable
0.42
Optional
0.42
He
0.42
POSITIVE LOGITS
ideas
0.83
creations
0.83
products
0.80
practices
0.79
discoveries
0.79
methodologies
0.79
designs
0.79
endeavors
0.79
innovations
0.78
successes
0.78
Activations Density 0.673%