INDEX

Explanations

development, nothing, work, be

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

-1.39

ry

-1.33

ron

-1.32

-1.27

で

-1.23

į

-1.23

ts

-1.22

 alors

-1.20

-1.18

POSITIVE LOGITS

しちゃ

1.27

Seems

1.25

ürgen

1.24

 один

1.24

vább

1.23

 bête

1.22

 seemed

1.20

しなかった

1.20

guruan

1.19

 chưa

1.16

Activations Density 0.115%