INDEX

Explanations

any CPU / Observer / ptr

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 anyone

-0.11

 anybody

-0.11

ANY

-0.10

OTA

-0.10

ANY

-0.10

edImage

-0.10

eenth

-0.10

 whatsoever

-0.09

olta

-0.09

ä»»

-0.09

POSITIVE LOGITS

 mang

0.09

wich

0.08

quiet

0.08

(.

0.08

one

0.08

 exception

0.08

uet

0.07

 Loud

0.07

lib

0.07

ou

0.07

Activations Density 0.036%

No Known Activations