INDEX

Explanations

for followed by a noun

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

-1.59

点此

-1.37

 insist

-1.33

 всю

-1.33

 पढ़ें

-1.31

 весь

-1.29

テール

-1.27

mapsto

-1.27

tadora

-1.27

 década

-1.24

POSITIVE LOGITS

the

1.58

饪

1.47

 maupun

1.47

鎂

1.45

 zvuky

1.38

oczes

1.38

</u>

1.38

 був

1.37

 znacznie

1.36

蜮

1.36

Activations Density 0.082%