INDEX

Explanations

step followed by number

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

-1.87

ቷ

-1.71

yyyyy

-1.56

老爷

-1.55

\*

-1.53



-1.53

 fleur

-1.51

-1.50

 ocasião

-1.48

 vader

-1.46

POSITIVE LOGITS

໌

2.16

ጺ

2.09

⌵

1.96

 рекомендуется

1.93

它的

1.86



1.84

ꪹ

1.83

 gleichen

1.82

 Для

1.81

<bos>

1.78

Activations Density 0.040%