INDEX

Explanations

start actions

based on training data

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 themselves

1.12

mselves

0.88

 हमसे

0.76

 ponerse

0.76

 తమ

0.71

 нам

0.69

给我们

0.65

 கொண்டனர்

0.65

 pihaknya

0.63

 nám

0.62

POSITIVE LOGITS

 myself

2.63

 خودم

1.87

 Myself

1.75

我自己

1.67

 نفسي

1.58

 دارم

1.47

хожу

1.47

 हूं

1.45

 છું

1.44

 بودم

1.36

Activations Density 0.842%