INDEX

Explanations

specific technical components

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

 Musée

-1.54

 данного

-1.49



-1.43

-1.42

 taches

-1.34

 جدید

-1.30

 pauv

-1.29

adır

-1.28

"$_

-1.27

ſy

-1.27

POSITIVE LOGITS

 that

1.87

can

1.76

 will

1.73

 always

1.60

may

1.55

as

1.54

 when

1.50

for

1.49

 Whether

1.38

 either

1.37

Activations Density 0.254%