INDEX

Explanations

"-ologically tested" or "pdf release"

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

 what

-1.98

the

-1.49

who

-1.38

but

-1.30

so

-1.27

 farb

-1.25

屋さん

-1.23

 perché

-1.20

いただき

-1.20

<eos>

-1.19

POSITIVE LOGITS

 veliko

1.41

richting

1.35

derung

1.34

らせて

1.34

发布时间

1.32

 izdel

1.31

;%%

1.29

ご予約

1.27

нт

1.23

 življen

1.23

Activations Density 0.225%