INDEX

Explanations

working with clients

The neuron fires on words naming or quantifying people or groups (e.g. clients, teams, students, “new,” “existing,” “multiple,” “many of,” etc.)—i.e. entities involved in collaborative or service contexts.

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

two

-1.89

in

-1.84

for

-1.64

 three

-1.61

 this

-1.43

 time

-1.42

情况下

-1.41

 most

-1.39

 even

-1.38

小行星

-1.38

POSITIVE LOGITS

蓠

1.54

◱

1.43

ihe

1.35

</h1>

1.33

timately

1.33

 delà

1.31

 يمكن

1.31

みな

1.29

 lancement

1.28

 Brésil

1.27

Activations Density 0.037%