INDEX

Explanations

institutions and groups

The neuron detects explicit numerical quantities or counts (e.g. numbers paired with entities like “five schools,” “130 schools,” “22 clubs,” etc.).

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

 obce

-1.56

 filma

-1.52

 perpé

-1.52

 maestra

-1.52

 fête

-1.50

 triom

-1.47

 pous

-1.45

 mola

-1.44

гато

-1.42

 camisetas

-1.42

POSITIVE LOGITS

to

2.73

who

2.33

 that

2.31

 whose

1.93

 which

1.80

 whom

1.62

th

1.54

habitude

1.52

 obyvateľov

1.51

 cuyos

1.50

Activations Density 0.069%