INDEX

Explanations

sentence-ending punctuation followed by a word

The neuron detects numeric tokens or digit sequences (e.g., years, section numbers, counts).

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

 ilustracji

-1.59

on

-1.52

in

-1.51

 epitom

-1.50

one

-1.37

 from

-1.36

 exquisitely

-1.35

by

-1.32

or

-1.31

 then

-1.31

POSITIVE LOGITS

andering

1.56

|}\

1.49

aneously

1.40

 potrebbe

1.38

annique

1.35

rizona

1.35

 pertinente

1.35

 .......

1.34

 semelh

1.34

 ſie

1.33

Activations Density 0.085%