INDEX

Explanations

foundational models, intelligence, DICOM, cling behavior, aid, machine learning, GUT, corruption

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

エア

1.00

^{*}(

0.99

休暇

0.93

 Lace

0.87

lou

0.85

チップ

0.85

 niece

0.84

翎

0.84

 nieces

0.84

ifl

0.83

POSITIVE LOGITS

now

1.69

 теперь

1.65

 kini

1.63

 ahora

1.60

 sekarang

1.59

 тепер

1.50

 kemudian

1.48

 потім

1.47

 then

1.42

 şimdi

1.42

Activations Density 0.357%