INDEX

Explanations

purpose or destination

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 zufried

-1.34

 einwand

-1.17

所有

-1.09

 mogelijkheid

-1.02

On

-1.02

 anoche

-1.01

It

-0.96

The

-0.96

 sanitarios

-0.95

 クリック

-0.95

POSITIVE LOGITS

 themselves

1.77

 their

1.71

themselves

1.45

 join

1.32

 here

1.19

 either

1.14

 сравнению

1.11

 ilyen

1.09

either

1.06

 these

1.05

Activations Density 0.024%

No Known Activations