INDEX

Explanations

asking for or requesting something

The neuron detects words and phrases that signal a request or call to action (e.g. “ask,” “asking,” “ask that,” “to…”).

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

 attaining

-1.47

 unruly

-1.38

 seemingly

-1.37

 changes

-1.37

 epitom

-1.35

vč

-1.35

將

-1.34

場

-1.34

各式

-1.33

各種

-1.31

POSITIVE LOGITS

 that

2.19

౭

1.84

almost

1.80

actually

1.77

unfortunately

1.74

those

1.71

generally

1.69

probably

1.69

 môžete

1.67

nearly

1.66

Activations Density 0.013%