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

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

Neuron 2315 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.6402. Min Act: -0.0000

Data Index: 796499 (C4 (Web Text))

Max Activating Token Index: 826

Text #1

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.6224. Min Act: -0.0000

Data Index: 603046 (C4 (Web Text))

Max Activating Token Index: 280

Text #2

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.6380. Min Act: -0.0000

Data Index: 774490 (C4 (Web Text))

Max Activating Token Index: 242

Text #3

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.6000. Min Act: -0.0000

Data Index: 397602 (C4 (Web Text))

Max Activating Token Index: 264

Text #4

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.5906. Min Act: -0.0000

Data Index: 36070 (C4 (Web Text))

Max Activating Token Index: 34

Text #5

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.5588. Min Act: -0.0000

Data Index: 114241 (C4 (Web Text))

Max Activating Token Index: 549

Text #6

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.5703. Min Act: -0.0000

Data Index: 355884 (C4 (Web Text))

Max Activating Token Index: 924

Text #7

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.5787. Min Act: -0.0000

Data Index: 507382 (C4 (Web Text))

Max Activating Token Index: 980

Text #8

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.5563. Min Act: -0.0000

Data Index: 45133 (C4 (Web Text))

Max Activating Token Index: 374

Text #9

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.5321. Min Act: -0.0000

Data Index: 1246383 (C4 (Web Text))

Max Activating Token Index: 343

Text #10

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.5086. Min Act: -0.0000

Data Index: 65759 (C4 (Web Text))

Max Activating Token Index: 58

Text #11

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.5211. Min Act: -0.0000

Data Index: 516652 (C4 (Web Text))

Max Activating Token Index: 993

Text #12

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.4882. Min Act: -0.0000

Data Index: 307933 (C4 (Web Text))

Max Activating Token Index: 1006

Text #13

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.4947. Min Act: -0.0000

Data Index: 150016 (C4 (Web Text))

Max Activating Token Index: 125

Text #14

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.4637. Min Act: -0.0000

Data Index: 859547 (C4 (Web Text))

Max Activating Token Index: 811

Text #15

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.4756. Min Act: -0.0000

Data Index: 370119 (C4 (Web Text))

Max Activating Token Index: 21

Text #16

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.4813. Min Act: -0.0000

Data Index: 615538 (C4 (Web Text))

Max Activating Token Index: 453

Text #17

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.4998. Min Act: -0.0000

Data Index: 831253 (C4 (Web Text))

Max Activating Token Index: 436

Text #18

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.4727. Min Act: -0.0000

Data Index: 87546 (C4 (Web Text))

Max Activating Token Index: 404

Text #19

Max Range: 0.6402. Min Range: -0.6402

Max Act: 0.4942. Min Act: -0.0000

Data Index: 276706 (C4 (Web Text))

Max Activating Token Index: 260