INDEX

Explanations

called or "so-called"

This is called

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

プ

0.44

著者

0.41

 ನಾಲ್

0.40

icel

0.40

zte

0.39

 നാല

0.38

 пяти

0.37

緑

0.37

聁

0.37

 Nature

0.37

POSITIVE LOGITS

 sogenannten

1.73

 sogenannte

1.66

 called

1.63

稱為

1.59

tzv

1.55

 sogen

1.55

 termed

1.54

called

1.52

 cosidd

1.51

称为

1.50

Activations Density 0.132%