INDEX

Explanations

respectively

The neuron activates on numeric tokens (integers or decimal numbers), i.e. any standalone numbers.

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

ings

-1.42

 would

-1.35

-1.27

かの

-1.23

厘米

-1.21

ské

-1.20

is

-1.19

 когда

-1.17

 waar

-1.16

空调

-1.16

POSITIVE LOGITS

 fühlt

1.41

umblr

1.40

 diferenças

1.39

About

1.37

 کردیم

1.35

colori

1.30

 richtet

1.29

嚓

1.28

 geschieht

1.27

cibo

1.27

Activations Density 0.015%