Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4725 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.2731. Min Act: -0.0002

Data Index: 1500084 (Python Code)

Max Activating Token Index: 250

Text #1

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.1326. Min Act: -0.0002

Data Index: 693741 (C4 (Web Text))

Max Activating Token Index: 490

Text #2

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.1500. Min Act: -0.0002

Data Index: 1577365 (Python Code)

Max Activating Token Index: 108

Text #3

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.2258. Min Act: -0.0002

Data Index: 1555258 (Python Code)

Max Activating Token Index: 149

Text #4

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.1797. Min Act: -0.0002

Data Index: 1086235 (C4 (Web Text))

Max Activating Token Index: 52

Text #5

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.2133. Min Act: -0.0002

Data Index: 1611630 (Python Code)

Max Activating Token Index: 57

Text #6

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.0449. Min Act: -0.0002

Data Index: 1417230 (Python Code)

Max Activating Token Index: 501

Text #7

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.1474. Min Act: -0.0002

Data Index: 1466492 (Python Code)

Max Activating Token Index: 187

Text #8

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.0311. Min Act: -0.0002

Data Index: 242539 (C4 (Web Text))

Max Activating Token Index: 650

Text #9

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.1440. Min Act: -0.0002

Data Index: 463367 (C4 (Web Text))

Max Activating Token Index: 10

Text #10

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.0872. Min Act: -0.0002

Data Index: 576051 (C4 (Web Text))

Max Activating Token Index: 104

Text #11

Max Range: 1.2731. Min Range: -1.2731

Max Act: 0.9966. Min Act: -0.0002

Data Index: 1411029 (Python Code)

Max Activating Token Index: 255

Text #12

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.1051. Min Act: -0.0002

Data Index: 1558507 (Python Code)

Max Activating Token Index: 36

Text #13

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.0734. Min Act: -0.0002

Data Index: 1584942 (Python Code)

Max Activating Token Index: 286

Text #14

Max Range: 1.2731. Min Range: -1.2731

Max Act: 0.9281. Min Act: -0.0002

Data Index: 1407627 (Python Code)

Max Activating Token Index: 853

Text #15

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.0374. Min Act: -0.0002

Data Index: 1075950 (C4 (Web Text))

Max Activating Token Index: 208

Text #16

Max Range: 1.2731. Min Range: -1.2731

Max Act: 0.9886. Min Act: -0.0002

Data Index: 142587 (C4 (Web Text))

Max Activating Token Index: 264

Text #17

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.0346. Min Act: -0.0002

Data Index: 984709 (C4 (Web Text))

Max Activating Token Index: 9

Text #18

Max Range: 1.2731. Min Range: -1.2731

Max Act: 0.9995. Min Act: -0.0002

Data Index: 1382100 (Python Code)

Max Activating Token Index: 189

Text #19

Max Range: 1.2731. Min Range: -1.2731

Max Act: 1.0511. Min Act: -0.0002

Data Index: 1137821 (C4 (Web Text))

Max Activating Token Index: 164