INDEX

Explanations

books and their creators

The neuron strongly activates on tokens typical of picture-book bibliographic entries—e.g. “picture,” “book(s),” page counts, publisher names, and related metadata.

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

 there

-1.70

_{

-1.49

-1.44

you

-1.41

↵

-1.39

-1.34

up

-1.22

-1.20

 там

-1.19

 levels

-1.16

POSITIVE LOGITS

selben

1.48



1.45

UGHS

1.41

푣

1.36

cepatan

1.35

 zlep

1.34

gewöhn

1.32

accor

1.32

">!

1.30

ducteur

1.29

Activations Density 0.179%