INDEX

Explanations

the followed by specific entities

This neuron detects the start of a definition-style sentence, especially the “The … is a” construction that introduces an entity.

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

to

-1.43

the

-1.29

<th>

-1.27

irmanship

-1.27

 همان

-1.22

 same

-1.16

Nå

-1.12

 colspan

-1.11

 Ensuite

-1.10

 приготовить

-1.08

POSITIVE LOGITS

his

1.70

 ње

1.45

in

1.38

了他的

1.35

ってみて

1.27

 hänen

1.27

of

1.23

Isra

1.20

 Berikut

1.17

文化的

1.17

Activations Density 0.244%