INDEX

Explanations

names from prefixes and suffixes

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ç§°

-0.09

ental

-0.09

vester

-0.08

 lust

-0.08

leston

-0.08

 deut

-0.08

Åĵ

-0.08

/Main

-0.08

Ø§Ø¡

-0.08

lust

-0.08

POSITIVE LOGITS

*'

0.08

 boosted

0.08

 visita

0.07

_Man

0.07

aben

0.07

ó

0.07

 Instit

0.07

.this

0.07

 Aero

0.07

apos

0.07

Activations Density 2.108%

No Known Activations