INDEX

Explanations

numbers, punctuation, and conjunctions

conditions regarding harm

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

 Typhoon

0.42

搔

0.42

रित

0.41

ider

0.41

لعاب

0.41

好像

0.40

rogram

0.40

tric

0.39

chrotron

0.39

 cyanide

0.39

POSITIVE LOGITS

Ip

0.49

pesar

0.48

were

0.47

 पार्वती

0.47

!=

0.46

cca

0.46

Seed

0.45

アカ

0.45

Á

0.45

0.44

Activations Density 0.001%