INDEX

Explanations

3D transformation and rotation

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Horizontal

-0.84

Rie

-0.82

 night

-0.77

一张

-0.74

 две

-0.71

Voor

-0.68

developers

-0.68

äger

-0.67

unggu

-0.67

Two

-0.66

POSITIVE LOGITS

pivot

1.02

 pivot

0.95

xyz

0.93

 local

0.91

center

0.91

 axis

0.90

xyz

0.90

 translation

0.87

 center

0.86

 axes

0.86

Activations Density 0.015%