INDEX

Explanations

ideal conditions

The neuron activates on numeric tokens (digits or multi‐digit sequences, like phone numbers and section numbers).

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

reathe

0.90

 capitalist

0.86

 billionaire

0.85

нской

0.84

im

0.82

 Walmart

0.81

сны

0.80

 NYPD

0.80

ны

0.79

 paperback

0.79

POSITIVE LOGITS

 tissus

0.91

的情况

0.79

 organes

0.77

 changements

0.76

 élément

0.76

 âge

0.75

實際

0.73

某些

0.73

 éléments

0.73

 possibilités

0.73

Activations Density 0.007%