INDEX

Explanations

describes things

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

entire

-1.18

➖

-1.02

 filmu

-1.01

☆☆☆☆☆

-0.98

۳

-0.98

 ошибки

-0.95

 konkret

-0.94

liegenden

-0.92

 elegante

-0.91

 compagnie

-0.91

POSITIVE LOGITS

 kind

1.87

 time

1.38

 kinds

1.23

guy

1.10

 seemingly

1.09

 young

1.03

 matter

1.03

Kind

1.00

 Kind

0.96

原因

0.94

Activations Density 0.007%