INDEX

Explanations

preposition + noun/determiner

The neuron strongly responds to common function words and auxiliaries that signal formal or passive‐voice constructions (e.g. by, in, as, from, to, be).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ſtra

-1.45

”。

-1.45

——”

-1.44

}}=

-1.41

 moſt

-1.41

 två

-1.40

”:

-1.38

蓠

-1.36

ahuila

-1.36

 supprimer

-1.36

POSITIVE LOGITS

 that

2.30

 while

1.93

 another

1.90

if

1.78

 with

1.75

 before

1.64

 make

1.63

 some

1.55

 perhaps

1.54

 primarily

1.53

Activations Density 0.667%