INDEX

Explanations

leaving or exiting

The neuron activates on words and phrases that signal departure or exit (e.g. “headed for the door,” “foot out the door,” “on his way out,” “packed up”).

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

 like

-1.59

 such

-1.48

 sicherzustellen

-1.38

and

-1.29

 akhirnya

-1.27

 старый

-1.23

*,

-1.20

 erreicht

-1.18

lliant

-1.16

 gezeigt

-1.16

POSITIVE LOGITS

 diſ

1.55

меров

1.41

unless

1.36

žio

1.33

leben

1.32

σίας

1.27

eker

1.27

 préoccup

1.26

ָר

1.26

 دیگر

1.26

Activations Density 0.057%