INDEX

Explanations

locations after "in"

The neuron detects prepositional location phrases—especially the word “in” when introducing a place name.

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

two

-2.02

Two

-1.77

of

-1.72

two

-1.53

 three

-1.52

-1.47

 наша

-1.46

his

-1.42

 from

-1.41

POSITIVE LOGITS

 cartier

1.48

一惊

1.23

 toco

1.23

峮

1.23

примеча

1.21

ars

1.20

zta

1.20

 ボール

1.17

TAT

1.16

 extraordinaire

1.16

Activations Density 0.056%