INDEX

Explanations

command or code start

The neuron activates on tokens that are names of functions or methods being called (i.e. identifiers immediately preceding a parenthesis).

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

蛱

-1.30

侂

-1.27

 były

-1.27

 inizia

-1.24

 evtl

-1.18

álny

-1.18

ooooooo

-1.16

臵

-1.16

!!!!!!!

-1.13

荭

-1.13

POSITIVE LOGITS

类的

1.13

崭

1.13

 принад

1.09

おもちゃ

1.07

ニーズ

1.07

 after

1.07

パートナー

1.07

 каже

1.06

ズニー

1.05

of

1.03

Activations Density 0.000%