INDEX

Explanations

for loop with counter i

The neuron fires on standalone numeric literals in code (e.g. loop bounds or constant values).

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

 even

-1.65

 from

-1.55

by

-1.43

 only

-1.36

 including

-1.25

for

-1.22

 also

-1.20

 filós

-1.20

 something

-1.19

-1.16

POSITIVE LOGITS

 vocale

1.30

sifs

1.21

 accesible

1.19

姈

1.17

möglichen

1.14

plätze

1.11

 vider

1.11

 camion

1.10

 conspic

1.09

upuncture

1.09

Activations Density 0.002%