INDEX

Explanations

The neuron lights up on tokens from formal scientific or mathematical notation—e.g. LaTeX delimiters and commands (“$”, “right”, “}”, “%”) and technical prefixes in chemistry (“photo”).

labels and specific entities

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

in

-1.84

by

-1.58

on

-1.48

to

-1.27

 being

-1.24

 from

-1.23

-1.22

などが

-1.22

is

-1.14

the

-1.13

POSITIVE LOGITS

Уже

1.45

Tils

1.41

Скачать

1.41

Вопрос

1.40

――――

1.38

Уважа

1.33

Тема

1.32

Faktor

1.29

Текст

1.27

もうすぐ

1.26

Activations Density 0.000%