INDEX

Explanations

numbers and chapter headings

The neuron fires on standalone numeric tokens or short sequences of digits (e.g., page/post numbers, scores, years, or docket identifiers).

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

Their

-1.59

 أيضا

-1.55

But

-1.50

Who

-1.44

Can

-1.35

 klikken

-1.34

也可以

-1.30

Also

-1.30

ти

-1.29

POSITIVE LOGITS

ซ

1.33

cardigan

1.31

ٞ

1.29

 Kasus

1.28

Nachteile

1.27

ὺς

1.27

Balas

1.26

Razer

1.24

karena

1.23

ﮯ

1.23

Activations Density 0.019%