INDEX

Explanations

This neuron detects numeric tokens (i.e. sequences of digits representing numbers).

know that

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

einer

-1.29

‫‪

-1.26

Waktu

-1.21

 которой

-1.20

휀

-1.19

Nilai

-1.17

Хочу

-1.17

 zdrow

-1.17

这么多

-1.15

Langkah

-1.11

POSITIVE LOGITS

of

1.27

但

1.24

It

1.22

他是

1.17

and

1.13

 about

1.12

 विवरण

1.11

but

1.06

THEY

1.06

Activations Density 0.067%