INDEX

Explanations

postal addresses and cities

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

 specifically

-0.98

at

-0.90

 during

-0.84

 because

-0.82

吧

-0.82

到

-0.78

 discovered

-0.77

 based

-0.77

ories

-0.76

在

-0.75

POSITIVE LOGITS

 mail

1.20

 sauve

1.10

 mailing

1.05

 mailbox

1.04

 email

1.02

 téléphone

1.02

 postal

0.98

 envelopes

0.95

 Postal

0.91

 handels

0.90

Activations Density 0.032%