INDEX

Explanations

referring to people or body

The neuron strongly activates on words referring to a human individual or their body (e.g. “person,” “body,” “him,” “them”).

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

 uniquement

-1.38

ling

-1.30

CCIÓN

-1.26

 seulement

-1.23

aaaaaa

-1.21

只

-1.21

 BATT

-1.20

zzz

-1.20

by

-1.18

only

-1.17

POSITIVE LOGITS

 Schauspielerin

1.73

 بدون

1.72

Они

1.55

งิน

1.54

 такие

1.45

 variously

1.43

Leírás

1.40

vč

1.39

Татьяна

1.38

hướng

1.38

Activations Density 0.055%