INDEX

Explanations

asking to rate or state preference

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Gives

1.11

"(

1.06

 Alerts

1.06

𝙣

1.04

 Yên

1.02

 документи

1.02

 Statements

1.02

":

1.01

 telev

1.01

POSITIVE LOGITS

three

1.48

三种

1.35

rim

1.30

 इत्यादि

1.27

など

1.27

 tiga

1.27

etc

1.26

 وغیرہ

1.26

linear

1.26

texture

1.26

Activations Density 0.001%