INDEX

Explanations

specific topics and clauses

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

<strong>

-1.20

].

-1.12

</sup>

-1.07

みたいに

-1.02

 porque

-1.01

 occorre

-0.97

 paraît

-0.97

 maîtres

-0.96

/**

-0.95

Omg

-0.95

POSITIVE LOGITS

 sowohl

1.29

 skall

1.11

 gelar

1.08

幹線

1.05

 MINE

1.05

 Hvad

1.04

 ۱۶

1.02

as

1.01

utiliz

1.00

 Einen

0.99

Activations Density 0.054%