INDEX

Explanations

pages and models

This neuron strongly activates on occurrences of the identifier “Pages” (as used in Razor Pages namespaces, directives, types, etc.).

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

="

-1.85

."

-1.82

-1.39

 نشده

-1.34

,"

-1.26

 самая

-1.25

 coun

-1.24

you

-1.23

.,

-1.23

 برابر

-1.20

POSITIVE LOGITS

ILITIES

1.55

“-

1.52

uelle

1.47

”-

1.46

딪

1.42

 Gespräche

1.41

 …”

1.38

際に

1.38

”،

1.38

 celebrar

1.37

Activations Density 0.002%