INDEX

Explanations

The neuron strongly activates on list-item markers (e.g. leading dashes “–” or asterisks “*”) that denote bullet points or YAML-style sequence entries.

list items or delimiters

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

or

-1.38

as

-1.34

 melap

-1.30

 like

-1.27

 向け

-1.23

unduh

-1.23

龔

-1.20

seamnă

-1.18

any

-1.16

 menik

-1.16

POSITIVE LOGITS

 breathtaking

1.60

}^{+}$

1.46

有意思

1.45

饪

1.42

1.41

...:

1.40

…’

1.38

,’”

1.38

 území

1.38

/**

1.36

Activations Density 0.005%