INDEX

Explanations

company prefixes

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orgeous

-0.07

ulous

-0.07

ipher

-0.07

izarre

-0.07

Wol

-0.06

ardon

-0.06

idious

-0.06

 latter

-0.06

icient

-0.06

ICH

-0.06

POSITIVE LOGITS

FAQ

0.16

 FAQs

0.16

Pru

0.12

 woes

0.12

 hacks

0.12

 podcast

0.12

poses

0.12

 Mens

0.11

 retrospective

0.11

 sampler

0.11

Activations Density 0.361%