INDEX

Explanations

section beginnings

This neuron fires on second-person and deictic function words (e.g. “you,” “this,” “if,” “because”) that signal direct address or explanatory/conditional statements.

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

 bądź

-1.01

či

-0.98

gnet

-0.96

xon

-0.93

udan

-0.90

 respective

-0.88

ike

-0.87

Ĉ

-0.86

odore

-0.86

\}.

-0.85

POSITIVE LOGITS

 koš

1.20

Exemples

1.08

有很多

1.05

Введение

1.00

Some

0.93

 birçok

0.92

 بعضی

0.90

┖

0.88

 iconLine

0.88

 gjerne

0.86

Activations Density 0.000%