INDEX

Explanations

sharing thoughts and wisdom

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

Its

-1.32

 بيان

-1.28

Instead

-1.26

 săptăm

-1.26

此外

-1.25

 gånger

-1.24

Specifically

-1.23

 allerlei

-1.22

 koste

-1.22

But

-1.20

POSITIVE LOGITS

1.45

PerformLayout

1.43

ating

1.38

boxylic

1.35

 AssemblyCompany

1.34

眼前的

1.30

archical

1.30

ramientas

1.27

这话

1.24

for

1.23

Activations Density 0.010%