INDEX

Explanations

making for

trained on usage starting

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

 démontré

0.54

ิง

0.53

ంగా

0.52

ങ്ങളിൽ

0.52

 comportement

0.50

 samen

0.49

 Marcell

0.48

eme

0.47

 scient

0.47

 erfolgreich

0.47

POSITIVE LOGITS

start

0.52

poke

0.52

pot

0.48

tron

0.47

frozen

0.46

pOt

0.46

path

0.45

pickup

0.45

venture

0.44

shakes

0.44

Activations Density 0.001%