INDEX

Explanations

call to action links

The neuron flags call-to-action link phrases (e.g. “Click HERE,” “Discover more,” “Find out more,” “Learn more below”) prompting the reader to navigate for additional information.

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

umerable

-1.70

 также

-1.55

 cũng

-1.45

也不

-1.45

сную

-1.41

 also

-1.38

сторія

-1.37

⎆

-1.31

க்கும்

-1.30

 same

-1.29

POSITIVE LOGITS

いましたが

1.48

of

1.40

 here

1.27

 before

1.23

MongoDB

1.20

 ранее

1.19

しましたが

1.18

늦

1.18

 ellos

1.16

してい

1.16

Activations Density 0.109%