INDEX

Explanations

names of people

The neuron is highly sensitive to named entities and titles—that is, it fires strongly on tokens that are parts of proper nouns (people’s names, film or album titles, organizations, etc.).

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

 éviter

-1.81

 ")";

-1.65

-1.59

中学生

-1.58

[{

-1.57

Instead

-1.55

 déterminer

-1.55

*/

-1.55

 GOING

-1.53

*/

-1.52

POSITIVE LOGITS

</strong>

1.74

 because

1.73

 muchacho

1.69

on

1.66

 From

1.58

in

1.55

尔的

1.53

 after

1.52

 during

1.52

bě

1.47

Activations Density 5.264%