INDEX

Explanations

exemption from rules

The main thing this neuron does is detect occurrences of the word “exemption.”

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

ть

-3.02

ar

-2.94

of

-2.52

茈

-2.50

 вернулся

-2.30

di

-2.28

icksburg

-2.22

all

-2.22

る

-2.19

in

-2.17

POSITIVE LOGITS

replyreport

2.38

 soñ

2.33

넋

2.30

м

2.28

絰

2.20

㉗

2.17

 大全

2.17

नलिखित

2.13

hacer

2.13

anleitung

2.11

Activations Density 0.003%