INDEX

Explanations

times and dates

The neuron specializes in picking out numeric date/time stamps (e.g. days, months, years, or clock times) in document metadata.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

if

-2.13

of

-2.05

 since

-1.81

 because

-1.66

for

-1.65

to

-1.63

has

-1.53

in

-1.52

‘

-1.50

 during

-1.45

POSITIVE LOGITS

 évit

1.90

蓐

1.73

跸

1.67

狁

1.63

 sientes

1.59

 transparentes

1.57

 modrá

1.57

 parfü

1.55

莼

1.53

髫

1.52

Activations Density 0.030%