INDEX

Explanations

intensive care unit

The neuron detects references to intensive care units (ICU), with particularly strong activation for the phrases "intensive care" and "care units."

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

tamaris

-1.83

我一直

-1.81

噥

-1.78

釙

-1.75

 geldig

-1.72

数を

-1.72

 pakan

-1.72

'])->

-1.70

vidéo

-1.66

 美しい

-1.65

POSITIVE LOGITS

1.71

1.58

www

1.51

we

1.50

見

1.45

$\

1.44

is

1.38

ètement

1.38

itigen

1.34

 course

1.34

Activations Density 0.027%