INDEX

Explanations

names and titles

This neuron fires on capitalized tokens that are part of proper names or titles (e.g. names of works, people, games, etc.).

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

-1.34

-1.22

-1.17

–

-1.14

-1.13

ским

-1.12

 pasar

-1.10

-1.05

她的

-1.03

-1.02

POSITIVE LOGITS

)}]{

1.48

 confiança

1.39

ież

1.36

 denselben

1.35

 than

1.35

 Gegensatz

1.28

inerario

1.25



1.23

痪

1.23

zwi

1.23

Activations Density 0.009%