INDEX

Explanations

horizontal separators

The neuron does not activate on any tokens here—i.e. it isn’t detecting any pattern in these excerpts.

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

to

-2.55

 giuri

-1.82

 still

-1.75

 this

-1.74

 with

-1.73

on

-1.72

 oddly

-1.72

 time

-1.62

 innumerable

-1.62

 ложки

-1.60

POSITIVE LOGITS

,”

1.98

opies

1.83

卺

1.75

 detalj

1.74

他们

1.70

vua

1.67

镡

1.66

hè

1.65

 از

1.59

،

1.59

Activations Density 0.092%