INDEX

Explanations

firmware and boot software

The neuron is detecting mentions of firmware or software/boot‐level components (e.g., “firmware,” “software,” “boot,” “coreboot,” “microprogrammed”).

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

न्दु

-1.41

are

-1.41

ğını

-1.39

 明星

-1.38

lemme

-1.34

 mnoho

-1.34

 litté

-1.34

ᐕ

-1.34

 labai

-1.32

 więc

-1.30

POSITIVE LOGITS

1.73

is

1.56

now

1.40

 there

1.40

1.33

 فهو

1.31

number

1.27

ーション

1.27

 работой

1.27

Activations Density 0.032%