INDEX

Explanations

member identifiers

The neuron fires on words and phrases referring to “members” or “membership” (e.g. affiliated members, member companies, membership counts).

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

↵

-1.59

if

-1.47

 about

-1.37

-1.36

</u>

-1.32

-1.30

we

-1.30

caid

-1.25

ца

-1.23

 said

-1.23

POSITIVE LOGITS

archiwizowane

1.42

 деца

1.42

 fidé

1.40

ﮦ

1.38

 siyang

1.37

xiaomi

1.33

 всичко

1.33

鸲

1.31

 SUCH

1.31

 μπορεί

1.30

Activations Density 0.083%