INDEX

Explanations

voices arguing that

The neuron activates on occurrences of the verb “argue” (in all its conjugated forms).

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

 ждать

-1.38

 assure

-1.32

’，

-1.29

酲

-1.24

 originate

-1.23

这意味着

-1.23

最后由

-1.22

部位

-1.18

与此同时

-1.18

メリカ

-1.16

POSITIVE LOGITS

 argues

1.80

 that

1.71

屺

1.52

 argued

1.50

mpagne

1.46

why

1.38

 whether

1.35

筇

1.31

ೣ

1.31

 SINCE

1.30

Activations Density 0.031%