INDEX

Explanations

linking HTML `<a>` tags and phrases from study

This neuron detects HTML anchor (“a”) elements (links) in the markup.

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

was

-1.69

 beschä

-1.38

ních

-1.32

.....

-1.31

其他

-1.30

 übrigen

-1.28

 demás

-1.27

 took

-1.24

did

-1.21

POSITIVE LOGITS

是怎么回事

1.48

1.42



1.33

 skak

1.31

डी

1.31

⫾

1.31

 vernac

1.30

1.29

erobic

1.28

 telo

1.27

Activations Density 0.002%