Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 9 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.2234. Min Act: -0.0006

Data Index: 1073951 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1912. Min Act: -0.0006

Data Index: 288084 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1832. Min Act: -0.0006

Data Index: 123170 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1799. Min Act: -0.0006

Data Index: 1161533 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1720. Min Act: -0.0006

Data Index: 1270657 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1726. Min Act: -0.0006

Data Index: 1233843 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1743. Min Act: -0.0006

Data Index: 730039 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1727. Min Act: -0.0005

Data Index: 1039108 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1709. Min Act: -0.0006

Data Index: 909402 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1683. Min Act: -0.0006

Data Index: 1324990 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1682. Min Act: -0.0006

Data Index: 1144175 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1666. Min Act: -0.0006

Data Index: 786112 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1666. Min Act: -0.0006

Data Index: 678861 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1660. Min Act: -0.0006

Data Index: 586603 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1700. Min Act: -0.0006

Data Index: 931450 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1626. Min Act: -0.0006

Data Index: 305633 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1551. Min Act: -0.0006

Data Index: 561715 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1593. Min Act: -0.0006

Data Index: 284052 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1587. Min Act: -0.0006

Data Index: 1280551 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.2234. Min Range: -0.2234

Max Act: 0.1520. Min Act: -0.0006

Data Index: 56725 (C4 (Web Text))

Max Activating Token Index: 1023