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

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

Neuron 3662 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3838. Min Act: -0.0001

Data Index: 590627 (C4 (Web Text))

Max Activating Token Index: 781

Text #1

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.4156. Min Act: -0.0001

Data Index: 1256502 (C4 (Web Text))

Max Activating Token Index: 865

Text #2

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3987. Min Act: -0.0001

Data Index: 1187234 (C4 (Web Text))

Max Activating Token Index: 454

Text #3

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3995. Min Act: -0.0001

Data Index: 834196 (C4 (Web Text))

Max Activating Token Index: 100

Text #4

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3645. Min Act: -0.0001

Data Index: 9903 (C4 (Web Text))

Max Activating Token Index: 557

Text #5

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3686. Min Act: -0.0001

Data Index: 301086 (C4 (Web Text))

Max Activating Token Index: 470

Text #6

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3540. Min Act: -0.0001

Data Index: 86223 (C4 (Web Text))

Max Activating Token Index: 345

Text #7

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3452. Min Act: -0.0001

Data Index: 581384 (C4 (Web Text))

Max Activating Token Index: 984

Text #8

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3355. Min Act: -0.0001

Data Index: 1206302 (C4 (Web Text))

Max Activating Token Index: 629

Text #9

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3661. Min Act: -0.0001

Data Index: 165635 (C4 (Web Text))

Max Activating Token Index: 84

Text #10

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3450. Min Act: -0.0001

Data Index: 9466 (C4 (Web Text))

Max Activating Token Index: 180

Text #11

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3471. Min Act: -0.0000

Data Index: 1401787 (Python Code)

Max Activating Token Index: 221

Text #12

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3354. Min Act: -0.0001

Data Index: 1236639 (C4 (Web Text))

Max Activating Token Index: 23

Text #13

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3425. Min Act: -0.0001

Data Index: 111730 (C4 (Web Text))

Max Activating Token Index: 995

Text #14

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3592. Min Act: -0.0001

Data Index: 303926 (C4 (Web Text))

Max Activating Token Index: 663

Text #15

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3264. Min Act: -0.0001

Data Index: 141114 (C4 (Web Text))

Max Activating Token Index: 506

Text #16

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3018. Min Act: -0.0001

Data Index: 485134 (C4 (Web Text))

Max Activating Token Index: 559

Text #17

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3191. Min Act: -0.0001

Data Index: 624677 (C4 (Web Text))

Max Activating Token Index: 33

Text #18

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3194. Min Act: -0.0001

Data Index: 1101195 (C4 (Web Text))

Max Activating Token Index: 391

Text #19

Max Range: 0.4156. Min Range: -0.4156

Max Act: 0.3289. Min Act: -0.0001

Data Index: 132203 (C4 (Web Text))

Max Activating Token Index: 386