INDEX

Explanations

'why did' joke setup

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

nomenclature

1.14

World

1.13

grouped

1.12

 nomenclature

1.11

 strewn

1.09

 varies

1.06

 comprehensive

1.05

Forecast

1.05

 sprawling

1.04

broad

1.04

POSITIVE LOGITS

 anschließend

1.65

 diciendo

1.65

 deoarece

1.61

 dizendo

1.57

 poiché

1.55

 afirmó

1.53

 swoim

1.51

 sodass

1.51

 namelijk

1.47

 alors

1.46

Activations Density 0.356%