INDEX

Explanations

accusing or complaining

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

 heed

-0.11

 obedient

-0.10

 listening

-0.10

 listens

-0.10

 Independent

-0.10

 awaited

-0.09

 heel

-0.09

 independent

-0.09

uzu

-0.09

Listeners

-0.09

POSITIVE LOGITS

 accusing

0.24

 condemn

0.22

 complaint

0.21

 confronting

0.21

 confront

0.21

 accuse

0.21

 condemnation

0.21

 complain

0.21

 accusation

0.21

 complaining

0.20

Activations Density 0.277%