INDEX

Explanations

rules and guidelines

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uni

-0.10

ESS

-0.09

 pent

-0.08

 pard

-0.08

pom

-0.08

orial

-0.08

 Funding

-0.08

uh

-0.08

EXT

-0.08

asso

-0.08

POSITIVE LOGITS

 enforced

0.20

 governing

0.20

 rules

0.20

 imposed

0.20

 stricter

0.19

/rules

0.19

 guidelines

0.18

 strict

0.18

 requirements

0.18

 standards

0.17

Activations Density 1.853%