INDEX

Explanations

Scattered, floating, standing, sits

The neuron fires on list-introducing or section-heading verbs (e.g. “Included,” “Provided,” “Also,” “Not included,” “Pictured,” etc.) that signal the introduction of items or content.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

}+(

0.61

ப்படவில்லை

0.60

 يجعل

0.60

 αφού

0.57

}}.

0.57

combined

0.57

िफाई

0.57

})\

0.56

addClass

0.56

*}

0.56

POSITIVE LOGITS

are

0.88

is

0.78

 هستند

0.77

 zijn

0.76

 were

0.74

 janë

0.74

 jsou

0.73

 löyty

0.71

 ovat

0.71

 komen

0.71

Activations Density 0.120%