INDEX

Explanations

timestamps and code diffs

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

-1.36

了一個

-1.34

://

-1.30

 história

-1.27

一辆

-1.24

mäßigen

-1.22

 食事

-1.20

覚醒

-1.18

('

-1.18

 však

-1.17

POSITIVE LOGITS

 your

1.94

our

1.77

we

1.72

 われ

1.70

 način

1.67

you

1.66

 splet

1.60

 najbol

1.54

我们将

1.52

 Drž

1.50

Activations Density 0.003%