INDEX

Explanations

logically contradicting statements

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

চারের

1.36

 bénéf

1.30

াস্ত

1.24

 Verbre

1.21

concrete

1.18

値

1.18

aktual

1.18

čního

1.17

ški

1.17

প্তি

1.15

POSITIVE LOGITS

 họp

1.84

 trại

1.79

vui

1.74

 सुबह

1.73

騁

1.71

 sông

1.71

arlene

1.70

 nghĩ

1.65

䒷

1.63

 halved

1.61

Activations Density 0.000%