INDEX

Explanations

connectors and articles

This neuron detects punctuation tokens that introduce or separate sections and questions—especially colons (:) and question marks (?)—as used in titles, headings, and interrogative sentences.

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

 from

-2.05

to

-1.63

in

-1.52

of

-1.42

for

-1.39

 after

-1.29

and

-1.20

 після

-1.20

 your

-1.17

 also

-1.17

POSITIVE LOGITS

 ceux

1.21

 adonan

1.16

 otros

1.16

žuje

1.16

 desbloquear

1.16

 persino

1.16

 других

1.15

 helicópter

1.15

1.14

ቷ

1.14

Activations Density 0.590%