INDEX

Explanations

code related to imports

This neuron primarily detects module import statements, especially the “from” keyword in import lines.

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

 kerap

-1.27

мость

-1.23

 that

-1.23

 смесь

-1.21

 zweiter

-1.20

 vengo

-1.20

 Mereka

-1.17

可以在

-1.17

 هذا

-1.16

سانی

-1.16

POSITIVE LOGITS

”？

1.39

⎚

1.29

its

1.23

куда

1.22

ynchronously

1.22

📺

1.20

 navig

1.20

:][

1.20

鈁

1.20

чнее

1.18

Activations Density 0.015%