INDEX

Explanations

difference between and

The neuron fires on the two key terms being compared in “Difference between X and Y” question titles.

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

Länge

-1.23

enaam

-1.23

 επίσης

-1.20

’,

-1.20

୫

-1.19

atellite

-1.16

skapet

-1.16

此外

-1.15

cheduler

-1.14

MessageOf

-1.14

POSITIVE LOGITS

 this

1.64

ceptable

1.53

 then

1.50

we

1.48

 than

1.40

 these

1.27

then

1.26

now

1.21

how

1.20

芋

1.20

Activations Density 0.071%