INDEX

Explanations

titles and ranks

The neuron spotlights formal titles and ranks (e.g. “Rear,” “Managing,” “Chief”) in text.

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

-2.27

 футболка

-1.34

 王子

-1.26

 ergonomics

-1.26

 speichern

-1.25

 Many

-1.25

 stammen

-1.25

Ѱ

-1.25

〤

-1.24

 انسانی

-1.23

POSITIVE LOGITS

↵↵

1.93

嘢

1.54

 honom

1.51

 ögon

1.48

 considère

1.41

 przygoto

1.39

任务

1.39

 flera

1.38

 plumme

1.35

 carcasa

1.34

Activations Density 0.069%