INDEX

Explanations

code blocks and function calls

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

 cinza

-1.26

 alcune

-1.21

 présenter

-1.19

 テープ

-1.13

 algumas

-1.12

ఽ

-1.10

⏑

-1.09

 feront

-1.08

 conexão

-1.08

 BECAUSE

-1.05

POSITIVE LOGITS

in

1.37

 from

1.34

or

1.30

 just

1.22

as

1.20

at

1.13

by

1.12

on

1.07

 with

1.02

 more

1.02

Activations Density 0.001%