INDEX

Explanations

New Zealand, Ukraine should

The neuron selectively activates on numeric expressions—particularly years, dates, and other number tokens.

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

に於

-1.34

 Talvez

-1.30

 evtl

-1.27

 ἐκ

-1.27

ܐ

-1.26

髋

-1.25

("/:

-1.24

jennifer

-1.20

 talvez

-1.20

 roślin

-1.16

POSITIVE LOGITS

ִּ

1.38

ּוֹ

1.34

־ה

1.31

ְּ

1.27

 γυναι

1.23

 Wartawan

1.18

וּ

1.18

 broadly

1.10

in

1.09

 arrangements

1.09

Activations Density 0.017%