INDEX

Explanations

year, date, numbers

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 authored

-0.22

@author

-0.17

 authors

-0.17

 author

-0.17

ä½ľèĢħ

-0.16

by

-0.15

 Authors

-0.14

-authored

-0.14

 Author

-0.14

Author

-0.14

POSITIVE LOGITS

hus

0.12

ardy

0.10

 butt

0.10

url

0.10

 Abstract

0.09

 runners

0.09

uant

0.09

Abstract

0.09

hare

0.09

 subt

0.09

Activations Density 1.240%