INDEX

Explanations

code comments and delimiters

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

်း

-0.82

ilesh

-0.82

 quelli

-0.73

だけの

-0.68

 Meets

-0.68

 Dukes

-0.68

教科

-0.67

 butcher

-0.66

ボトル

-0.65

 daran

-0.65

POSITIVE LOGITS

likler

0.78

EXO

0.77

LOO

0.75

BECK

0.72

葉

0.72

asiswa

0.72

AllAfrica

0.71

ატ

0.71

 Attributes

0.71

Suggestion

0.70

Activations Density 0.074%