INDEX

Explanations

closing code blocks

This neuron fires primarily on Ruby’s `end` keyword (i.e. the closing of blocks or definitions).

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

 Well

-0.95

farers

-0.92

伯爵

-0.84

жды

-0.83

;}

-0.82

guides

-0.80

iu

-0.78

 Namely

-0.77

 Zelanda

-0.77

̚

-0.77

POSITIVE LOGITS

end

1.52

 unless

1.12

 конце

1.05

 конец

1.00

 koniec

0.93

 końca

0.92

除非

0.91

 self

0.86

エンド

0.86

unless

0.86

Activations Density 0.010%