INDEX

Explanations

technical code or explanation

The neuron is triggered by numeric constants—particularly floating‐point numbers with decimal points.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 didSet

0.23

 json

0.23

 yaml

0.22

())

0.22

 substring

0.22

 tooltip

0.21

 $("#

0.21

 verbose

0.21

 offsetX

0.21

==

0.21

POSITIVE LOGITS

industrie

0.24

agricultural

0.19

reliant

0.18

 преимущественно

0.18

ړو

0.18

engineered

0.17

<unused622>

0.17

 अल्को

0.17

<unused621>

0.17

fring

0.17

Activations Density 0.006%