INDEX

Explanations

numbers eleven and twelve

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

For

-0.95

for

-0.88

For

-0.73

Hes

-0.71

 för

-0.70

ে

-0.69

 Starling

-0.68

forto

-0.68

FOR

-0.67

estia

-0.66

POSITIVE LOGITS

Twelve

0.82

eleven

0.81

 Eleventh

0.80

Eleven

0.79

 Eleven

0.75

 eleventh

0.74

 eleven

0.71

twelve

0.71

 Eller

0.71

 misspelled

0.71

Activations Density 0.466%