INDEX

Explanations

categorization

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

 triple

-0.11

ece

-0.10

anten

-0.10

Ups

-0.10

ailer

-0.09

Éµ

-0.09

zet

-0.09

ispens

-0.09

-inner

-0.09

POSITIVE LOGITS

 typically

0.11

 usually

0.10

aka

0.09

 exempl

0.09

 characterized

0.09

 most

0.09

 Usually

0.09

usually

0.09

 Typically

0.09

 generally

0.08

Activations Density 0.191%