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

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

Neuron 537 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: 0.3683. Min Range: -0.3683

Max Act: 0.3663. Min Act: -0.0002

Data Index: 840067 (C4 (Web Text))

Max Activating Token Index: 63

Text #1

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.3683. Min Act: -0.0002

Data Index: 1007153 (C4 (Web Text))

Max Activating Token Index: 283

Text #2

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2809. Min Act: -0.0002

Data Index: 696303 (C4 (Web Text))

Max Activating Token Index: 114

Text #3

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.3103. Min Act: -0.0002

Data Index: 9085 (C4 (Web Text))

Max Activating Token Index: 100

Text #4

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.3299. Min Act: -0.0002

Data Index: 666447 (C4 (Web Text))

Max Activating Token Index: 66

Text #5

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.3167. Min Act: -0.0002

Data Index: 311973 (C4 (Web Text))

Max Activating Token Index: 331

Text #6

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2015. Min Act: -0.0002

Data Index: 384675 (C4 (Web Text))

Max Activating Token Index: 217

Text #7

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2507. Min Act: -0.0002

Data Index: 532370 (C4 (Web Text))

Max Activating Token Index: 740

Text #8

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.3080. Min Act: -0.0002

Data Index: 464339 (C4 (Web Text))

Max Activating Token Index: 467

Text #9

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2859. Min Act: -0.0002

Data Index: 1297026 (C4 (Web Text))

Max Activating Token Index: 735

Text #10

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.3045. Min Act: -0.0002

Data Index: 46267 (C4 (Web Text))

Max Activating Token Index: 6

Text #11

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.3307. Min Act: -0.0002

Data Index: 1294326 (C4 (Web Text))

Max Activating Token Index: 139

Text #12

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2672. Min Act: -0.0002

Data Index: 62569 (C4 (Web Text))

Max Activating Token Index: 92

Text #13

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2435. Min Act: -0.0002

Data Index: 365575 (C4 (Web Text))

Max Activating Token Index: 220

Text #14

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2805. Min Act: -0.0002

Data Index: 781362 (C4 (Web Text))

Max Activating Token Index: 435

Text #15

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2692. Min Act: -0.0002

Data Index: 549143 (C4 (Web Text))

Max Activating Token Index: 506

Text #16

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.3024. Min Act: -0.0002

Data Index: 276300 (C4 (Web Text))

Max Activating Token Index: 806

Text #17

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2664. Min Act: -0.0002

Data Index: 527212 (C4 (Web Text))

Max Activating Token Index: 111

Text #18

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2350. Min Act: -0.0002

Data Index: 1209309 (C4 (Web Text))

Max Activating Token Index: 737

Text #19

Max Range: 0.3683. Min Range: -0.3683

Max Act: 0.2355. Min Act: -0.0002

Data Index: 251516 (C4 (Web Text))

Max Activating Token Index: 653