INDEX

Explanations

Exp $

This neuron detects isolated curly‐brace tokens (empty `{}` brace pairs) in code.

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

to

-1.52

by

-1.47

 from

-1.33

on

-1.32

in

-1.20

one

-1.16

for

-1.15

and

-1.09

 then

-1.07

-1.06

POSITIVE LOGITS

 posibilidades

1.25

 échou

1.16

 refroid

1.13

的金

1.13

 accél

1.12

 récup

1.11

 grueso

1.11

פורט

1.09

изображение

1.09

 możesz

1.09

Activations Density 0.026%