INDEX

Explanations

until

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 flesta

-0.70

 igång

-0.66

 écrits

-0.65

 écoute

-0.61

 varandra

-0.61

 nôtre

-0.59

 penduduk

-0.56

 arbete

-0.55

 vägen

-0.55

 gratuites

-0.54

POSITIVE LOGITS

 birth

0.69

 bone

0.66

 kasarigan

0.65

 brain

0.64

 nahilalakip

0.64

 death

0.62

 tissue

0.59

 pregnancy

0.57

 bones

0.57

EndContext

0.57

Activations Density 0.006%