INDEX

Explanations

paralleling shapes or structures

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

èĢĮéĻį

-0.08

emun

-0.08

yms

-0.08

ushi

-0.08

orte

-0.08

éļĺ

-0.07

reden

-0.07

ickle

-0.07

emp

-0.07

yens

-0.07

POSITIVE LOGITS

paralle

0.09

 paralle

0.08

 property

0.07

è¥Ħ

0.07

nesia

0.07

witz

0.07

Property

0.07

åıĺå½¢

0.07

brook

0.06

 displacement

0.06

Activations Density 0.001%