INDEX

Explanations

presenting information or circumstances

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

wie

2.00

ぎ

1.85

দ্য

1.80

wel

1.75

्‍य

1.67

</td>

1.65

ides

1.64

ائف

1.61

znacz

1.61

 آز

1.60

POSITIVE LOGITS

dSample

2.64

ेक्षित

2.55

allclasses

2.50

 Scalar

2.45

емых

2.43

ッジ

2.43

 régulièrement

2.35

 vassal

2.34

锏

2.33

 isotropic

2.32

Activations Density 0.888%