INDEX

Explanations

programming keywords and system constants

This neuron detects uppercase time‐unit tokens and duration labels (e.g. “HOUR,” “MIN,” “MONTH,” “YEAR,” “HOURS,” “MIN”).

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

any

-2.02

for

-2.02

our

-1.98

 only

-1.95

at

-1.74

and

-1.70

or

-1.63

 that

-1.62

by

-1.62

-1.59

POSITIVE LOGITS

 bisogna

1.70

 gonzález

1.63

 ſol

1.62

 célè

1.55

wurst

1.50

咂

1.49

 plafon

1.48

 oublier

1.48

 neceſſ

1.48

Its

1.47

Activations Density 0.159%