INDEX

Explanations

admit that / admit to

admit and acknowledge

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

க

0.61

 Ensure

0.59

ड

0.59

 Dona

0.58

ку

0.56

гов

0.56

ص

0.55

كان

0.54

ون

0.54

NMC

0.54

POSITIVE LOGITS

 admit

1.17

 admitted

1.16

 admitting

1.10

承认

1.09

 admits

1.01

 conceding

0.96

 admite

0.93

 admission

0.90

 admissions

0.89

 призна

0.88

Activations Density 0.037%