INDEX

Explanations

d3.js visualization elements

The neuron flags technical terms and identifiers from D3 code (e.g. method names like call, axis, force, directed, numeric “3,” etc.), effectively spotting D3‐related code snippets.

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

⼈

-1.35

or

-1.30

 }));

-1.30

 where

-1.25

 wherein

-1.23

élène

-1.23

($(

-1.23

”？

-1.23

ází

-1.22

 others

-1.21

POSITIVE LOGITS

</b>

1.41

lichte

1.16

UTELY

1.15

њи

1.14

 Before

1.13

 değiştir

1.13

rodz

1.12

teil

1.12

 vacate

1.11

 управление

1.09

Activations Density 0.012%