INDEX

Explanations

part of

This neuron responds to occurrences of the substring “set,” i.e. words beginning with “set.”

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

 refroid

-1.80

鸰

-1.80

ネー

-1.71

メロン

-1.69

シャレ

-1.66

 coû

-1.59

Ꭸ

-1.55

カシ

-1.52

 perfeita

-1.49

 rempli

-1.48

POSITIVE LOGITS

 also

2.25

for

2.16

 about

1.72

of

1.63

祖国

1.57

as

1.54

 ormai

1.52

文章

1.52

now

1.51

dzić

1.51

Activations Density 0.005%