INDEX

Explanations

past and present

The neuron detects contrastive or paired adjectives—especially temporal or spatial antonyms (e.g. “past / present,” “new / old,” “north / south”) in titles or headings.

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

-1.59

in

-1.46

-1.43

 your

-1.36

↵

-1.30

 podían

-1.28

-1.20

 meditación

-1.16

 frecuentemente

-1.13

 často

-1.12

POSITIVE LOGITS

 pretende

1.34

ULTS

1.19

peut

1.19

紅茶

1.19

 culmin

1.18

 suro

1.16

)";

1.16

 notori

1.15

 decidedly

1.14

ESH

1.14

Activations Density 0.050%