INDEX

Explanations

code keywords and names

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

 peines

-0.93

 proposent

-0.93

 rajut

-0.91

 vengo

-0.90

 didik

-0.87

拍了拍

-0.86

͗

-0.86

 konkrete

-0.85

素敵

-0.85

судар

-0.85

POSITIVE LOGITS

if

1.38

or

1.15

1.09

 only

1.03

return

1.00

 return

0.94

and

0.91

adren

0.89

 assess

0.87

 under

0.83

Activations Density 0.001%