INDEX

Explanations

status codes 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

allenge

-0.83

 where

-0.82

vartis

-0.81

NDICE

-0.81

廬

-0.80

 revelations

-0.79

 révèle

-0.78

addContainerGap

-0.78

走行

-0.78

 revelation

-0.77

POSITIVE LOGITS

 success

1.50

success

1.48

 successful

1.31

 sucesso

1.30

 Success

1.23

SUCCESS

1.20

successful

1.20

Success

1.16

 SUCCESS

1.16

 code

1.15

Activations Density 0.005%