INDEX

Explanations

model: identifying AI identity

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

1.55

foo

1.34

 potty

1.31

 grandma

1.30

 crappy

1.29

south

1.26

1.25

 Buffett

1.24

town

1.20

 smelly

1.18

POSITIVE LOGITS

 ৩৪

1.77

}_{{\

1.57

 ৩৮

1.57

 ৩৬

1.56

 ৩৭

1.55

 ৩৫

1.54

 ৬৮

1.52

 ৩৩

1.48

 समन्व

1.48

 দ্যা

1.47

Activations Density 1.004%