INDEX

Explanations

harmful material, bacteria, toxins, programs

harmful content

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

for

1.08

was

1.05

大

1.04

Type

0.98

 Datensch

0.98

问题

0.96

 Datenschutz

0.95

是不

0.93

的路

0.91

POSITIVE LOGITS

ین

1.19

 harmful

1.13

روں

1.09

ів

1.06

 harming

1.05

ล

1.01

ין

1.00

 detriment

0.98

 harmon

0.98

し

0.96

Activations Density 0.055%