INDEX

Explanations

questions and queries

The neuron responds to “open‐class” content words—nouns, verbs, and adjectives—while ignoring function words.

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

to

-1.70

so

-1.64

 such

-1.51

 that

-1.35

 every

-1.34

 this

-1.32

 homicidio

-1.28

 estos

-1.27

 these

-1.27

 each

-1.25

POSITIVE LOGITS

␮

1.29

 Etik

1.27

₰

1.24

 cikin

1.22

 revend

1.21

ᕱ

1.17

 siècles

1.16

 vuitton

1.15



1.15

"]="

1.13

Activations Density 0.000%