INDEX

Explanations

referencing prior information

text elements and structure

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

 ৫০০

0.42

 रिजर्व

0.41

 animais

0.40

 পশু

0.39

 включают

0.38

ण्यासाठी

0.38

 দরজা

0.38

𝑂

0.38

 машины

0.38

 risco

0.38

POSITIVE LOGITS

 above

1.15

 выше

1.07

上述

1.05

 उपरोक्त

1.05

above

1.04

 powy

1.04

上記の

1.02

上記

1.01

 aforesaid

0.91

 acima

0.91

Activations Density 0.110%