INDEX

Explanations

closing parenthesis and conjunction

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

-2.91

-1.59

-1.40

<>

-1.34

（

-1.30

 مؤرشف

-1.27

呖

-1.26

歺

-1.26

 enfin

-1.24

изнь

-1.21

POSITIVE LOGITS

and

2.03

、「

1.60

(!__

1.59

 (!__

1.46

ventajas

1.42

 arşivlendi

1.38

 iNdEx

1.32

inerario

1.32

Nei

1.31

’.

1.30

Activations Density 0.257%