INDEX

Explanations

film, animation, actors, illustrations

The neuron fires on named entities—proper nouns like artist or album names, song/video titles, and familiar section‐heading terms.

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

</h2>

-1.53

–

-1.42

-1.25

junit

-1.24

 such

-1.21

);

-1.21

-1.19

 dětí

-1.19

.<

-1.18

restrial

-1.16

POSITIVE LOGITS

conocimiento

1.24

mejores

1.23

información

1.20

 spesies

1.20

sly

1.17

nunca

1.17

máscara

1.17

características

1.16

capulco

1.16

Eles

1.16

Activations Density 0.399%