INDEX

Explanations

lists names and locations

principles and concepts

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

 asesinato

0.56

 いつ

0.55

 extermin

0.55

 atrocities

0.55

 imati

0.54

 asesin

0.53

 avait

0.52

 incentivize

0.52

 Nası

0.52

 heinous

0.51

POSITIVE LOGITS

ics

0.59

erv

0.55

mx

0.54

border

0.50

hound

0.50

pl

0.48

bs

0.47

bx

0.47

ethers

0.46

Activations Density 0.000%