INDEX

Explanations

pronoun followed by verb or possessive

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

 also

-2.41

of

-2.33

and

-1.89

Also

-1.80

its

-1.66

 said

-1.59

 также

-1.48

or

-1.40

 նաև

-1.34

 then

-1.31

POSITIVE LOGITS

 verschillende

1.38



1.38

QLabel

1.38

 många

1.37

OMING

1.36

 flera

1.33

gebraucht

1.32

 meil

1.27

NewUrl

1.27

énez

1.27

Activations Density 0.077%