INDEX

Explanations

how things appear

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

场比赛

-1.00

 autonomia

-0.98

 dinam

-0.96

 notori

-0.95

 consistente

-0.93

 carcasa

-0.93

本质

-0.92

 régal

-0.91

 pecuni

-0.91

ceous

-0.91

POSITIVE LOGITS

 like

2.05

 innoc

1.35

to

1.35

innoc

1.20

 daunting

1.19

 harmless

1.16

 insignificant

1.14

 simple

1.12

 trivial

1.09

 logical

1.09

Activations Density 0.038%