INDEX

Explanations

processes and outcomes

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

 that

-2.03

 which

-1.67

to

-1.54

 ambao

-1.31

Still

-1.27

 different

-1.22

機會

-1.21

为什么会

-1.21

 notably

-1.20

Usually

-1.20

POSITIVE LOGITS

of

2.97

 jajaja

1.40

韭

1.23

 違い

1.21

 zapato

1.21

1.19

kaza

1.16

воде

1.16

нул

1.13

atau

1.13

Activations Density 0.226%