INDEX

Explanations

things happen

This neuron responds to numeric tokens and measurements (e.g. standalone numbers and numerical values).

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

為に

-1.37

えっ

-1.31

たびに

-1.21

凄く

-1.18

のかも

-1.17

わけで

-1.16

ネットで

-1.16

ようになって

-1.13

ごめん

-1.13

 ziemlich

-1.12

POSITIVE LOGITS

參考文獻

1.25

 even

1.19

if

1.16

 just

1.15

、

1.14

 when

1.13

 that

1.06

now

1.05

:^{

1.03

fah

1.02

Activations Density 0.007%