INDEX

Explanations

file paths and code

The neuron responds to code‐style file path fragments—i.e. directory or filename tokens containing forward-slash separators.

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

женская

-1.37

iembre

-1.29

hradun

-1.27

蜃

-1.27

tapete

-1.23

/^

-1.23

ctory

-1.23

lü

-1.21

leather

-1.20

spanish

-1.16

POSITIVE LOGITS

1.57

1.47

 that

1.46

ないです

1.42

 attempts

1.37

誰が

1.30

 autorytatywna

1.30

 fordert

1.30

没能

1.29

how

1.28

Activations Density 0.012%