INDEX

Explanations

names and capitalized words

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

 when

-1.35

 even

-1.14

 there

-1.14

When

-1.02

He

-0.99

 When

-0.95

 their

-0.94

when

-0.93

Hi

-0.92

hi

-0.91

POSITIVE LOGITS

Shy

0.99

Hydroxy

0.96

 Huang

0.93

 hydroxy

0.91

hup

0.90

 hund

0.89

 Hutchison

0.89

HttpResponse

0.89

 Howell

0.87

 Humphrey

0.86

Activations Density 0.012%