INDEX

Explanations

Mr or Ms titles

The neuron detects personal honorific titles (e.g. “Mr.” or “Ms.”) preceding names.

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

-1.85

 Sedangkan

-1.60

所有

-1.60

 want

-1.59

 Bentuk

-1.59

 wybrać

-1.53

 Kedua

-1.53

 Meskipun

-1.53

 möglichen

-1.52

 robić

-1.48

POSITIVE LOGITS

一顿

1.75

</sub>

1.63

 dança

1.60

</i>

1.58

 óleo

1.52

齶

1.48

 نوج

1.43

一辆

1.42

一段时间

1.41

变了

1.41

Activations Density 0.017%