INDEX

Explanations

hardware and its properties

The neuron fires on mentions of hardware, i.e. it detects occurrences of the token “hardware.”

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

’

-2.28

ти

-2.11

et

-1.89

↵

-1.84

𐑂

-1.75

cznego

-1.70

い

-1.69

</td>

-1.68

-1.64

“

-1.62

POSITIVE LOGITS

 mejorando

1.89

 mengirimkan

1.86

镚

1.80

קות

1.80

嚶

1.78

 классов

1.76

 pomoci

1.74

 odpově

1.72

 zlep

1.72

 vollständ

1.71

Activations Density 0.014%