INDEX

Explanations

technical terms and their attributes

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

 éte

-2.52

หรือ

-2.27

 éton

-2.17



-2.11

 éta

-2.11

 trám

-2.06

 améli

-2.03

 déchir

-2.03

 refroid

-2.02

の方に

-2.02

POSITIVE LOGITS

is

2.31

 with

2.28

in

2.28

by

2.20

as

2.20

has

2.19

 after

2.16

can

2.13

at

2.00

1.95

Activations Density 0.022%