INDEX

Explanations

universal healthcare and basic income

New Auto-Interp

Top Features by Cosine Similarity

Configuration

Prompts (Dashboard)

10,000 prompts, 128 tokens each

Dataset (Dashboard)

lmsys/lmsys-chat-1m

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 democrat

-0.12

 dereg

-0.11

ifact

-0.10

 Newtown

-0.09

 Darwin

-0.09

ifecycle

-0.09

Fla

-0.09

 subsid

-0.09

 typically

-0.08

 oxid

-0.08

POSITIVE LOGITS

 Universal

0.22

Universal

0.20

 Basic

0.18

 unconditional

0.15

 Guaranteed

0.15

 guaranteed

0.14

Basic

0.14

ubi

0.14

 universal

0.14

Jub

0.13

Activations Density 0.061%