INDEX

Explanations

describing the aim or goal

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

":["

1.43

<strong>

1.38

 people

1.36

1.32

achers

1.31

down

1.30

ly

1.28

1.26

（

1.26

【

1.26

POSITIVE LOGITS

 героя

1.70

 суу

1.67

 অনে

1.65

кансер

1.65

 верно

1.64

 ڈویل

1.64

 hydraz

1.62

 பேரா

1.61

 герой

1.61

 правой

1.60

Activations Density 0.609%