INDEX

Explanations

introduces information or entities

The neuron fires strongly on numeric or quantitative tokens (e.g. counts, measurements, section‐heading numbers) in a document.

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

-1.66

 when

-1.55

 since

-1.34

的一些

-1.29

 there

-1.27

 здесь

-1.27

тися

-1.26

 this

-1.22

on

-1.22

sipasi

-1.21

POSITIVE LOGITS

its

1.58

 restera

1.44

1.43

っています

1.28

 inga

1.21

ってます

1.20

たって

1.18

 trois

1.17

 resistentes

1.17

('

1.16

Activations Density 0.177%