INDEX

Explanations

time durations

The neuron activates on numeric expressions—numbers or number-related words indicating counts or durations (e.g. “a few months,” “over a year”).

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

 beina

-1.55

is

-1.41

yesha

-1.40

CRIPTION

-1.36

Ꞓ

-1.34

expandindo

-1.32

眚

-1.31

логии

-1.30

 أكبر

-1.28

turning

-1.27

POSITIVE LOGITS

）」

1.38

 many

1.34

 đã

1.33

 gesamten

1.32

de

1.30

 permitiendo

1.29

 several

1.27

or

1.27

 oppure

1.25

↵

1.23

Activations Density 0.019%