INDEX

Explanations

The neuron flags mentions of professional personnel and their credentials or training (e.g. “staff,” “certified,” “technicians,” “experts,” “engineers,” etc.).

skilled professionals and experts

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

 after

-2.08

for

-1.81

 before

-1.78

two

-1.71

or

-1.71

 three

-1.69

by

-1.68

 while

-1.67

but

-1.64

 first

-1.52

POSITIVE LOGITS

who

2.13

 berger

1.52

 assurent

1.47

 basili

1.46

 rivi

1.45

 prenn

1.45

 boute

1.41

 tulum

1.41

 véhicules

1.40

 ananas

1.34

Activations Density 0.040%