INDEX

Explanations

used to describe or following descriptions

no pressure

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 journaling

0.59

 businesswoman

0.55

 Sonos

0.52

 GIRL

0.51

ressa

0.50

 magnétique

0.50

 Phyllis

0.50

 streetwear

0.50

℅

0.50

czą

0.49

POSITIVE LOGITS

ter

0.48

0.46

pleasant

0.44

を経て

0.44

unknown

0.44

continue

0.44

지고

0.43

Malay

0.42

0.41

Eastern

0.41

Activations Density 0.000%