INDEX

Explanations

create a project

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 however

1.20

 regardless

1.19

 contemplated

1.14

 disappro

1.13

 persuaded

1.10

 notwithstanding

1.08

 whether

1.08

 nevertheless

1.07

 불구하고

1.06

 considering

1.06

POSITIVE LOGITS

èmes

1.50

 गाड़ियों

1.41

Declar

1.38

Entities

1.34

țional

1.33

declar

1.28

ței

1.26

 पेड़

1.25

đi

1.24

Adress

1.22

Activations Density 0.009%