INDEX

Explanations

uniformly distributed

This neuron flags adverbial qualifiers ending in “-ly,” especially words like “uniformly” or “categorically.”

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

↵

-1.80

However

-1.61

 However

-1.38

 their

-1.30

Concerning

-1.27

 There

-1.24

 oddly

-1.24

Regarding

-1.23

 secas

-1.16

who

-1.13

POSITIVE LOGITS

borderSide

1.57

 både

1.53

吭

1.52

}))

1.50

 }));

1.47

alties

1.45

》.

1.45

...@

1.42

()',

1.41

tores

1.41

Activations Density 0.062%