INDEX

Explanations

driverAcademyAutonomousnewsPython

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

 Edition

-0.09

LM

-0.08

arat

-0.08

 Empire

-0.07

Eid

-0.07

 Gunn

-0.07

 Bingo

-0.07

LM

-0.07

Har

-0.07

 scram

-0.07

POSITIVE LOGITS

 deprived

0.11

bay

0.10

ITS

0.10

aar

0.10

 provoked

0.10

esch

0.10

eer

0.10

 Augustine

0.10

 Hust

0.10

CHAIN

0.10

Activations Density 0.001%