INDEX

Explanations

sending

The neuron detects quantitative mentions of items or people being “sent” or “sending,” i.e. numeric amounts tied to the verb send.

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

寕

-2.28

-2.11

 kleineren

-1.83

 wußte

-1.82

 ähnlichen

-1.78

 vinculados

-1.72

 เขา

-1.71

有多少

-1.71

 conocida

-1.69

ाउनलोड

-1.68

POSITIVE LOGITS

Our

2.22

縯

1.97

Cuáles

1.96

 onze

1.88

跸

1.88

lij

1.83

那时

1.78

賸

1.78

//});

1.76

%)

1.75

Activations Density 0.022%