INDEX

Explanations

very short introduction

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

 LUMP

2.91

тьми

2.78

 Dangerous

2.76

ुर्

2.75

𝘤

2.70

ccia

2.67

merksamkeit

2.67

िस्ट

2.64

ನ್ನು

2.60

くださ

2.55

POSITIVE LOGITS

sville

2.40

্রেট

2.36

ie

2.34

<h4>

2.20

ITECT

2.17

eur

2.13

רץ

2.12

dsale

2.10

ždy

2.09

메

2.09

Activations Density 0.001%