INDEX

Explanations

roman numerals in historical contexts

The neuron is detecting numeric tokens (e.g. years and other multi‐digit 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

");

-1.60

 Abgerufen

-1.41

 песен

-1.39

-1.37

ға

-1.36

та

-1.34

醗

-1.34

喔

-1.33

");

-1.31

%");

-1.31

POSITIVE LOGITS

of

1.57

IDATE

1.22

 vodou

1.21

𝐊

1.21

 About

1.20

 voyons

1.20

gabung

1.10

 skut

1.08

 ähnliche

1.06

 gewöhn

1.02

Activations Density 0.029%