INDEX

Explanations

acknowledgment or introduction

This neuron detects acknowledgment language, particularly mentions of reviewers, feedback, and expressions of thanks.

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

idak

-1.20

it

-1.17

and

-1.17

żdy

-1.13

not

-1.13

 wund

-1.13

之以

-1.13

槻

-1.12

違和

-1.12

引用元

-1.10

POSITIVE LOGITS

 całym

1.34

\}$

1.27

Προ

1.23

え

1.21

 dateFormat

1.21

úč

1.20

 dalších

1.19

كن

1.19

เหมือน

1.18

 diğer

1.16

Activations Density 0.010%