INDEX

Explanations

sub- word followed by of

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

isn

-1.71

-1.56

ovať

-1.55

 ส์

-1.46

 tři

-1.45

絝

-1.45

 ardından

-1.45

精力

-1.43

weren

-1.43

skosten

-1.41

POSITIVE LOGITS

of

2.09

the

1.80

 within

1.78

to

1.77

1.69

 else

1.64

。

1.60

媜

1.54

 cantora

1.52

 only

1.51

Activations Density 0.043%