INDEX

Explanations

identifying pairs

This neuron responds to occurrences of the coordinating word “both.”

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

 idé

-1.28

˅

-1.22

TWO

-1.18

onu

-1.16

ندق

-1.16

 even

-1.15

ㅠ

-1.14

 extraordin

-1.12

டும்

-1.10

лії

-1.09

POSITIVE LOGITS

are

1.52

 этих

1.30

 jeweils

1.23

 trámites

1.21

らしいです

1.16

 verwendet

1.15

1.12

 shown

1.09

Activations Density 0.033%