INDEX

Explanations

no guaranteed single magic

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

1.91

 कुछ

1.89

 Some

1.72

娅

1.69

 There

1.68

्स

1.66

sá

1.66

 However

1.64

iac

1.63

də

1.63

POSITIVE LOGITS

strlen

1.86

രംഭ

1.83

>−</

1.76

Donnell

1.70

す楽

1.69

pmatrix

1.67

త్మ

1.63

 toán

1.63

嘢

1.62

ethyl

1.60

Activations Density 0.090%