INDEX

Explanations

for the sake of

The neuron fires on words that signal a purpose or intent statement (e.g. “for the sake of,” “in anticipation of,” “to,” “providing,” “improving”).

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

for

-2.22

in

-2.06

on

-2.02

of

-1.91

off

-1.84

 like

-1.80

to

-1.79

 ways

-1.78

way

-1.77

as

-1.74

POSITIVE LOGITS

 konstant

1.73

 rafraî

1.70

 déliv

1.70

 rafra

1.66

 refroid

1.59

oooo

1.56

 konfig

1.54

 corrom

1.52

 dokumenti

1.52

oooooo

1.48

Activations Density 0.188%