INDEX

Explanations

blog post introduction or filing

This neuron detects tokens related to document metadata—words and numbers used in publication bylines, dates, and citation details.

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

 Abon

-1.15

 Erbe

-1.14

一会儿

-1.13

 this

-1.09

وء

-1.07

電話番号

-1.07

líben

-1.06

 rutina

-1.05

=*

-1.03

its

-1.02

POSITIVE LOGITS

of

1.41

是怎么回事

1.22

1.08

ského

1.02

</em>

1.02

が

1.02

 лв

1.01

ıyordu

1.01

 لأنه

1.01

rt

1.01

Activations Density 0.000%