INDEX

Explanations

the start of enumerated lists

The neuron activates on “Table of Contents” (and similar “Contents”) headings, i.e. markers of a document’s table‐of‐contents section.

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

by

-1.19

it

-1.16

any

-1.14

as

-1.10

 most

-1.09

 those

-1.06

However

-1.06

 either

-1.06

 However

-1.03

 many

-1.02

POSITIVE LOGITS

ｇ

1.08

introduction

1.05

まず

1.04

Zunanje

0.97

 cremallera

0.96

!==

0.94

 collè

0.94

$\

0.94

 Tienen

0.93

<>

0.93

Activations Density 0.021%