INDEX

Explanations

analysis and vulnerabilities

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

עם

2.02

 cons

1.97

<%=

1.97

Republican

1.94

 परसेंट

1.93

1.92

 proviso

1.89

హి

1.84

St

1.83

ление

1.83

POSITIVE LOGITS

vā

2.43

ﻒ

2.40

فت

2.37

狽

2.37

ून्य

2.31

uania

2.30

ત્મ

2.23

ویل

2.19

umia

2.18

发动机

2.15

Activations Density 0.152%