INDEX

Explanations

election information and dates

The neuron fires on capitalized named‐entity tokens—especially place names and political/administrative units.

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

you

-1.47

 abgerufen

-1.27

 Prozent

-1.24

 ۲۰۱

-1.22

erweise

-1.16

 exquisite

-1.16

isn

-1.11

={

-1.11

 Zugriff

-1.10

 Create

-1.09

POSITIVE LOGITS

 späteren

1.73

 السياس

1.51

После

1.40

 actionTypes

1.39

 dunklen

1.27

igth

1.22

这两

1.21

饼

1.20

Etter

1.18

Mål

1.18

Activations Density 0.001%