INDEX

Explanations

noun phrases and titles

The neuron is triggered by proper names and formal titles—i.e. named entities such as persons, places, organizations, or official roles.

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

-1.51

 meraih

-1.19

しくない

-1.18

 گیا

-1.16

に行

-1.05

 ਨਾਲ

-1.05

 așa

-1.05

ِّ

-1.03

ere

-1.01

たちに

-1.00

POSITIVE LOGITS

new

1.54

ór

1.22

 nieuwe

1.20

 femininos

1.15

 whil

1.13

también

1.09

新しい

1.09

\\

1.08

 nuevas

1.08

 также

1.08

Activations Density 0.219%