INDEX

Explanations

a or an

The neuron fires on words that introduce a method or approach—tokens like “a,” “an,” “using,” “different,” “nontraditional,” “individualised,” etc., that typically kick off a methodological phrase.

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

in

-1.38

’

-1.31

 when

-1.28

 where

-1.21

at

-1.13

 places

-1.09

-1.07

窺

-1.02

剧照

-1.02

 '../../../../

-1.02

POSITIVE LOGITS

 erreur

1.41

+}\

1.28

 olie

1.23

ptt

1.18

komt

1.16

 berge

1.16

 heng

1.13

…»

1.12

illotine

1.12

ҟ

1.11

Activations Density 0.092%