INDEX

Explanations

refusal of 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

No Comments

Negative Logits

 Amazon

0.82

 एप्पल

0.81

Amazon

0.80

 அலெ

0.80

MEL

0.80

 अमेजन

0.77

 Phrase

0.74

եցին

0.74

amazon

0.73

Apple

0.73

POSITIVE LOGITS

↵↵↵↵↵↵↵↵↵

1.20

↵↵↵↵↵↵

1.16

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.13

↵↵↵↵↵

1.08

cimentos

1.06

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.04

↵↵↵↵↵↵↵↵↵↵↵

1.04

↵↵↵↵↵↵↵

1.04

↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.03

↵↵↵↵↵↵↵↵↵↵↵↵

1.02

Activations Density 0.032%