INDEX

Explanations

normal or historical state

The neuron detects adverbs (and similar phrases) that indicate habitual or customary frequency (e.g., “usually,” “normally,” “as usual”).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

Idioma

0.66

小米

0.64

 संस्करण

0.64

 Từ

0.64

DLLEXPORT

0.64

öt

0.63

̀

0.63

zeitig

0.63

ಿಸ್ಟ

0.63

설

0.62

POSITIVE LOGITS

 Normally

1.36

 normally

1.24

Normally

1.23

 suele

1.15

 suelen

1.13

normally

1.10

 normalerweise

1.10

 normalmente

1.09

 szok

1.05

 عادة

1.03

Activations Density 0.397%