INDEX

Explanations

code assignment and structures

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

-1.98

/*

-1.89

for

-1.84

-1.76

),

-1.73

»,

-1.72

They

-1.70

-1.68

or

-1.66

-1.64

POSITIVE LOGITS

ſſen

2.02

 plazos

2.00

 ajustar

1.89

桄

1.81



1.77

 same

1.76

へ向

1.71

 trámites

1.71

 beiden

1.70

 visualizar

1.70

Activations Density 0.026%