INDEX
Explanations
This neuron fires on various forms of the word “present” (e.g. presented, presents).
New Auto-Interp
Negative Logits
Mojo
-0.07
ávací
-0.06
log
-0.06
elk
-0.06
Shift
-0.06
(dc
-0.06
authToken
-0.06
Fork
-0.06
Eff
-0.06
(ic
-0.06
POSITIVE LOGITS
presented
0.15
Presented
0.13
presenting
0.12
presentation
0.12
Presents
0.11
present
0.11
presents
0.11
Presentation
0.10
-present
0.10
présent
0.10
Activations Density 0.052%