INDEX

Explanations

pike where, Perfect Can, Computing devices

This neuron detects place names ending with the suffix “-ville.”

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

 choroby

-1.28

 こと

-1.20

 hanya

-1.19

 produktu

-1.19

 semakin

-1.16

بيت

-1.15

 ścian

-1.15

例如

-1.14

 vín

-1.13

цыі

-1.13

POSITIVE LOGITS

里有

1.27

if

1.15

 vincula

1.14

 появился

1.13

若干

1.12

 urgently

1.11

دارة

1.09

 irresistible

1.08

 zależ

1.06

 contempla

1.05

Activations Density 0.000%