INDEX

Explanations

asking "what is"

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

 بشكل

2.04

 meaningful

1.91

 optimal

1.82

 paperwork

1.81

 impactful

1.77

 proactive

1.71

 বেশিরভাগ

1.67

 problematic

1.67

 groundbreaking

1.67

 basics

1.65

POSITIVE LOGITS

 ибо

1.47

ſ

1.39

 whereupon

1.31

 whence

1.30

 thereupon

1.27

And

1.24

And

1.22

 forthwith

1.21

 doubtless

1.20

 thence

1.20

Activations Density 0.351%