INDEX

Explanations

anisotropy or directional properties

New Auto-Interp

Configuration

Dataset (Dashboard)

Various

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Fizz

-0.08

 coined

-0.08

 ابت

-0.08

amu

-0.08

 Dawson

-0.07

raise

-0.07

 dwarf

-0.07

Au

-0.07

台

-0.07

halve

-0.07

POSITIVE LOGITS

 directional

0.13

(Direction

0.13

Directional

0.12

方向

0.12

 anis

0.12

 orientations

0.12

-direction

0.12

 direction

0.11

 orientation

0.11

 directions

0.11

Activations Density 0.008%