INDEX

Explanations

names after punctuation

The neuron selectively responds to tokens that are part of named entities or formal references—particularly person names, bracketed citations, and other proper nouns.

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

</strong>

-1.77

 because

-1.50

 usually

-1.46

 gelingt

-1.41

</h4>

-1.35

-1.34

 probably

-1.34

 often

-1.32

 typically

-1.32

 entstehen

-1.29

POSITIVE LOGITS

who

2.09

bloomberg

1.67

who

1.62

 जिन

1.46

1.45

又は

1.41

jewish

1.41

并

1.40

 whom

1.39

 地图

1.39

Activations Density 0.069%