INDEX

Explanations

hyperlinks and urls

This neuron responds to individual digit tokens that are part of longer numeric sequences (e.g. IDs, timestamps, or multi‐digit codes).

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

but

-1.49

 could

-1.47

申し上げます

-1.35

 would

-1.24

 flesta

-1.23

josé

-1.21

strtotime

-1.18

 säga

-1.16

 сентября

-1.13

겉

-1.12

POSITIVE LOGITS

1.38

1.29

 erwähnten

1.20

 húmedo

1.18

 gesuchten

1.18

 angep

1.16

 bogs

1.14

 MATTERS

1.13

depends

1.13

 nachher

1.12

Activations Density 0.022%