INDEX

Explanations

810s

The neuron spikes on occurrences of the digit “1” in numeric strings (e.g. phone numbers, patent numbers, addresses).

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

sona

-0.90

Ủ

-0.87

!("

-0.86

 melhor

-0.81

！】

-0.79

 best

-0.79

)}$.

-0.77

 vários

-0.76

 another

-0.75

chauffer

-0.75

POSITIVE LOGITS

Trending

0.86

 видит

0.81

Discipl

0.81

(_.

0.79

财政

0.79

DOES

0.78

0.77

낵

0.77

jublj

0.76

νον

0.75

Activations Density 0.010%