INDEX

Explanations

health and wellness

The neuron activates on occurrences of the word “health” (and very closely related well-being terms).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

나

1.36

クシー

1.21

ש

1.18

המ

1.16

ך

1.16

RON

1.13

 sunroof

1.11

вропей

1.11

োধ

1.10

POSITIVE LOGITS

 iban

1.34

 आरोग

1.31

ine

1.30

PACT

1.19

 salud

1.15

wellness

1.14

 здоровья

1.13

 sanitaria

1.13

sein

1.10

 Gesund

1.10

Activations Density 0.044%