INDEX

Explanations

quotes or specific phrases

The neuron detects standalone numeric tokens (sequences of digits appearing as isolated numbers in the text).

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

at

-2.47

in

-2.44

for

-2.09

 with

-2.08

on

-2.08

or

-1.99

 three

-1.95

 just

-1.91

two

-1.86

 four

-1.85

POSITIVE LOGITS

<blockquote>

2.70

<strong>

2.25

メガネ

1.85

 isolamento

1.83

1.82

튿

1.80

糁

1.79

甗

1.75

<b>

1.69

訁

1.68

Activations Density 0.013%