INDEX

Explanations

seem adj

The neuron activates on evaluative adjectives or adverbs that express subjective judgment (e.g. “trivial,” “simple,” “practical,” “luxury”).

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

is

-1.41

 talaga

-1.24

llllll

-1.24

ÇÃO

-1.20

мето

-1.18

lllllllll

-1.16

cún

-1.14

TestingModule

-1.13

isierung

-1.12

queryInterface

-1.11

POSITIVE LOGITS

you

1.77

but

1.56

 même

1.55

釹

1.48

 sogar

1.46

 nawet

1.41

or

1.29

还不是

1.28

 bahkan

1.27

 olabilir

1.25

Activations Density 0.071%