INDEX

Explanations

guides

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

guides

-1.23

 Guides

-1.22

Guides

-1.05

 guides

-0.91

 rospy

-0.64

AutoScaleMode

-0.60

 guías

-0.57

quakes

-0.56

guards

-0.56

board

-0.50

POSITIVE LOGITS

bootstrapcdn

0.62

حياته

0.52

 powinny

0.50

 عنه

0.50

 vuelto

0.50

いった

0.49

 Parigi

0.48

 voul

0.48

ivano

0.48

umbered

0.47

Activations Density 0.024%