INDEX

Explanations

this. followed by object property

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

already

-1.63

 voegen

-1.56

昙

-1.55

 japonés

-1.55

 Scritto

-1.48

bezpiec

-1.48

cribes

-1.47

Mô

-1.45

herself

-1.45

купить

-1.41

POSITIVE LOGITS

↵

2.38

</h4>

2.02

↵↵↵

1.99

is

1.93

↵↵↵↵

1.88

 need

1.84

</em>

1.78

↵↵

1.76

}{

1.72

">

1.65

Activations Density 0.002%