INDEX

Explanations

reducing amount, pain, or stress

reducing size or scope

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

 rangkaian

0.62

 inflater

0.61

 apocalypse

0.59

 ajouter

0.58

 aquifers

0.58

 executar

0.57

 Situation

0.57

 clerks

0.55

 invertible

0.54

 postcards

0.54

POSITIVE LOGITS

reduce

1.13

 reduce

1.03

Reduce

1.01

 Reduce

0.96

Reducing

0.94

減

0.94

reduced

0.93

reduction

0.93

reducing

0.93

 lessened

0.89

Activations Density 0.064%