INDEX

Explanations

surnames

This neuron detects proper-person surnames (family names) in the text.

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

 DELLE

-1.13

ところに

-1.12

 parfüm

-1.07

「

-1.05

 bakteri

-1.05

θο

-1.03

関連記事

-1.03

 journalistes

-1.02

ようで

-1.01

gün

-1.01

POSITIVE LOGITS

日を

1.14

が

1.06

を

1.01

曲は

0.98

時間も

0.96

is

0.95

0.93

et

0.93

itp

0.92

時間は

0.90

Activations Density 0.034%