INDEX

Explanations

code variable assignments and function calls

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

itados

-1.45

-1.41

 PUSTAKA

-1.38

↵

-1.34

”-

-1.29

 Dinas

-1.29

-1.27

diens

-1.27

 веке

-1.26

 same

-1.25

POSITIVE LOGITS

 this

1.77

 fossero

1.53

 haberse

1.49

 descansar

1.48

 jemals

1.46

 continúa

1.43

 odası

1.38

 numerosas

1.37

 unusually

1.37

 estaban

1.37

Activations Density 0.008%