INDEX

Explanations

self and individual concepts

The neuron fires on words naming people or groups—that is, nouns and pronouns referring to individuals or populations.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

្នាំ

1.05

 capito

1.05

 popol

1.02

Ciao

1.01

 parten

1.00

['

0.96

 underside

0.95

chmod

0.95

bcrypt

0.94

 ispod

0.94

POSITIVE LOGITS

л

1.40

 sendiri

1.36

ständig

1.34

1.32

м

1.30

лично

1.26

SELF

1.25

ગત

1.25

 self

1.23

सेवी

1.22

Activations Density 2.670%