INDEX

Explanations

learning and gaining knowledge

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 putative

1.71

 supplic

1.68

纤

1.63

 pedig

1.61

দু

1.58

arr

1.52

 Branson

1.52

 shady

1.52

Bid

1.51

 elegant

1.50

POSITIVE LOGITS

 realizations

1.59

elernt

1.51

 learnings

1.46

 solidified

1.37

 consapevole

1.36

 internalized

1.35

fahrung

1.35

 realization

1.27

 آگ

1.26

烙

1.26

Activations Density 0.162%

No Known Activations