INDEX

Explanations

World at war

The neuron activates primarily on numeric tokens (digits and numbers).

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

 XXXI

-1.30

 XXXIV

-1.27

寻常

-1.25

 XXXVII

-1.24

 LXXX

-1.18

nî

-1.16

 XXXVI

-1.16

 XXXV

-1.15

ttes

-1.14

-1.13

POSITIVE LOGITS

at

1.47

has

1.38

 варіан

1.23

茨城

1.21

 with

1.21

 opción

1.19

 result

1.16

one

1.15

 leads

1.15

而在

1.13

Activations Density 0.131%