INDEX

Explanations

introduces what follows

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

そも

-1.13

ちょっとした

-1.12

 entweder

-1.02

あー

-1.01

 BECAUSE

-1.01

以外にも

-0.99

ちょこ

-0.96

んでしょう

-0.94

 selec

-0.92

 ____________

-0.92

POSITIVE LOGITS

in

1.23

at

1.13

 miljø

1.12

 brugt

1.11

 billede

1.06

 though

1.05

 millioner

1.04

 virker

1.01

 kuni

1.00

 medlems

0.98

Activations Density 0.131%