INDEX

Explanations

numbers and units

The neuron selectively activates on numeric characters or standalone number tokens.

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

of

-1.31

 which

-1.13

in

-1.05

or

-1.03

for

-1.03

:"-"`

-1.02

お待ちしております

-0.98

amblea

-0.93

でございます

-0.91

POSITIVE LOGITS

1.13

 cerimonia

1.00

Mafia

0.99

其余

0.99

 shewn

0.98

'/',

0.93

 tandis

0.91

 altres

0.91

 Synag

0.89

 autres

0.88

Activations Density 0.121%