INDEX

Explanations

by no means

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

ónimo

-0.81

setto

-0.81

curso

-0.81

lebih

-0.81

 MethodInfo

-0.80

tuigen

-0.75

 APPEAL

-0.74

Tahun

-0.73

 Stateful

-0.73

 grö

-0.73

POSITIVE LOGITS

 means

2.22

means

1.77

 stretch

1.45

 Means

1.41

Means

1.27

 MEANS

1.22

 означает

1.16

 mean

1.13

stretch

1.13

 bedeutet

1.12

Activations Density 0.008%