INDEX

Explanations

character encoding and programming

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

and

2.07

The

1.95

 from

1.93

 with

1.93

From

1.86

以及

1.84

 towards

1.82

и

1.78

 અને

1.76

and

1.73

POSITIVE LOGITS

 laatste

2.25

 junge

2.04

 meisten

2.01

 gesamte

1.96

 plupart

1.87

 meeste

1.86

 letzte

1.84

 fleste

1.81

 juiste

1.79

 latter

1.78

Activations Density 0.111%