INDEX

Explanations

news channels with numbers

This neuron detects local broadcast news station identifiers—network names, call letters, channel numbers, and affiliate tags.

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

 maakte

-1.27

one

-1.27

 evidently

-1.25

by

-1.22

Surprisingly

-1.19

<i>

-1.16

 that

-1.13

TWO

-1.13

也會

-1.13

POSITIVE LOGITS

zną

1.55

">'

1.49

臯

1.45

тите

1.43

绡

1.42

 Dinas

1.31

">-

1.30

物を

1.30

 pittores

1.30

>"

1.29

Activations Density 0.034%