INDEX

Explanations

expressions of surprise or unexpectedness

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

ArgsConstructor

-0.57

=$((

-0.57

dAtA

-0.54

 funky

-0.53

 weird

-0.51

odd

-0.50

 peculiar

-0.50

 evocative

-0.50

نين

-0.49

Encyklopedia

-0.47

POSITIVE LOGITS

 surprise

3.39

 Surprise

3.05

surprise

3.00

 surprised

2.92

Surprise

2.80

 surprises

2.73

surprised

2.47

 sorpresa

2.42

 surpresa

2.28

 Überras

2.23

Activations Density 0.023%