INDEX

Explanations

purpose of this

powers, culture

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

our

0.45

 přij

0.44

 jednoduch

0.44

to

0.44

 whiskey

0.44

 semplice

0.44

 officers

0.43

 puffs

0.43

 arrivals

0.42

 juxtap

0.41

POSITIVE LOGITS

ScienceStudent

0.55

 网络

0.52

lı

0.50

途径

0.50

हाना

0.48

trifluoromethyl

0.48

 کشت

0.48

 Idha

0.48

۱۲

0.47

łym

0.47

Activations Density 0.001%