INDEX

Explanations

code and special characters

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

oriously

-0.29

gunakan

-0.25

rippling

-0.25

rvine

-0.24

 mÃ¢

-0.24

swire

-0.23

rips

-0.23

 vidÃ©

-0.21

 encount

-0.21

 Leakage

-0.21

POSITIVE LOGITS

åıĸåĨ³

0.25

ä½»

0.25

æĢİä¹ĪåĽŀäºĭ

0.22

Ð¿Ð¾ÑĩÑĤ

0.22

åĶ·

0.21

}elseif

0.21

ä½ľåĵģåĨħå®¹

0.21

å¿ħä¸įåı¯

0.20

ä»İæł¹æľ¬

0.20

YPES

0.20

Activations Density 0.388%