INDEX

Explanations

must have + verb/adjective

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

 jusqu

-1.59

probably

-1.57

 généralement

-1.55

 ممکن

-1.41

 การ

-1.41

 pasaj

-1.41

་

-1.41

Ą

-1.40

 pude

-1.40

likely

-1.40

POSITIVE LOGITS

of

1.88

 with

1.81

 then

1.69

 which

1.67

for

1.66

 especially

1.62

to

1.60

 more

1.54

 added

1.50

 that

1.48

Activations Density 0.015%