INDEX

Explanations

Q:

This neuron fires on numeric tokens (digits, multi-digit numbers, and decimal points).

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

 would

-1.52

for

-1.23

 paraíso

-1.19

 versie

-1.14

gén

-1.13

 ratón

-1.12

setPower

-1.11

rure

-1.10

#__

-1.09

 Werken

-1.09

POSITIVE LOGITS

𓈒

1.27

bkz

1.25

瞜

1.20

今回も

1.16

 сообщил

1.15

ૉ

1.14

MATER

1.10

 tekrar

1.05

 médicament

1.05

it

1.03

Activations Density 0.000%