INDEX

Explanations

quantities or comparisons

The main thing this neuron does is detect quantitative expressions—numbers and words that quantify amounts (e.g. “one,” “five,” “every,” “any,” “more,” etc.).

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

ที่เรา

0.59

所

0.59

我們

0.56

 особенностей

0.56

ন্যায়

0.55

 గురించి

0.54

可以说是

0.54

 دیکھا

0.53

 మూడు

0.53

 বেঁচে

0.53

POSITIVE LOGITS

 kilograms

0.60

0.59

or

0.55

 liters

0.52

handedly

0.52

 amount

0.51

(=

0.51

and

0.50

 grams

0.49

volle

0.49

Activations Density 0.704%