INDEX

Explanations

without permission or consent

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

 dign

-0.12

uth

-0.09

 kindly

-0.08

 disobed

-0.08

GN

-0.08

 dangers

-0.08

 unwanted

-0.08

 spilled

-0.08

 unhealthy

-0.08

dere

-0.08

POSITIVE LOGITS

 proper

0.18

 consent

0.17

 permission

0.16

 prior

0.14

 adequate

0.13

 authorization

0.13

proper

0.13

 Hust

0.13

 Proper

0.13

 sufficient

0.12

Activations Density 0.066%