INDEX

Explanations

paired items or concepts

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

↵↵

-2.53

-2.42

↵

-2.30

-2.22

-2.03

ため

-1.95

 был

-1.80

{}

-1.77

<em>

-1.77

 должна

-1.71

POSITIVE LOGITS

阋

2.09

 masculinos

2.08

本章

2.05

 thier

1.90

 ansiedad

1.90

of

1.88

the

1.87

趼

1.87

窠

1.86

 indulge

1.81

Activations Density 0.004%