INDEX

Explanations

require path

This neuron detects parts of module import or require path strings, especially file names and directory names (often prefixed with “_”) in require('./…') calls.

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

 while

-1.27

 this

-1.20

 nestled

-1.16

had

-1.15

 utaf

-1.13

 displayed

-1.11

 kufanya

-1.11

 lipca

-1.07

 будущем

-1.06

 habari

-1.05

POSITIVE LOGITS

 cães

1.25

EVERY

1.23

Numéro

1.21

Especially

1.20

 espes

1.20

 interesses

1.18

discussed

1.15

Préparation

1.14

especially

1.10

 utilis

1.09

Activations Density 0.005%