INDEX

Explanations

"the" followed by technical nouns

The neuron activates on numeric tokens denoting decimal or floating-point values.

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

 …..

-1.84

 ……

-1.79

brtc

-1.70

GRAFIA

-1.70

'"

-1.64

cences

-1.61

 ‘’

-1.57

пление

-1.55

;&

-1.54

thmus

-1.49

POSITIVE LOGITS

at

2.13

 your

1.97

 both

1.86

 each

1.82

 these

1.78

for

1.74

 their

1.72

 this

1.70

of

1.59

all

1.59

Activations Density 0.142%