INDEX

Explanations

a place or facility

The neuron flags numeric expressions—especially multi‐digit numbers and decimals.

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

 anahtar

-1.55

 since

-1.53

if

-1.46

 that

-1.42

días

-1.38

 while

-1.35

 people

-1.35

-1.34

to

-1.32

 påvir

-1.32

POSITIVE LOGITS

équipements

1.41

 ders

1.30

 sitios

1.29

mie

1.28

 prins

1.23

 построен

1.23

比べると

1.20

 offens

1.20

起来的

1.20

THERE

1.18

Activations Density 0.061%