INDEX

Explanations

rowspan, direction, Hyper, Tech

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

 terke

-2.16

 ayakkab

-2.08

 bidra

-2.03

ített

-1.85

 område

-1.82

佷

-1.80

กรณ์

-1.80

畚

-1.79

tifying

-1.77

полне

-1.77

POSITIVE LOGITS

 rowspan

2.45

</strong>

2.34

2.09

of

2.06

ৄ

1.98

1.91

 jeweils

1.85

="

1.70

 четыре

1.66

 spä

1.64

Activations Density 0.000%