INDEX

Explanations

code endings, imports

The neuron fires most strongly on import‐ or require‐style module‐loading lines in code.

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

to

-1.23

 only

-1.16

를

-1.15

on

-1.07

 during

-1.01

 will

-0.99

 these

-0.95

が

-0.95

 small

-0.94

at

-0.93

POSITIVE LOGITS

 suoi

1.30

 его

1.01

 그의

1.00

 jego

0.98

 カット

0.97

its

0.95

 তার

0.94

 RECOMEND

0.93

 cinquième

0.92

edad

0.91

Activations Density 0.042%