Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 2020 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0486. Min Act: -0.0001

Data Index: 501470 (C4 (Web Text))

Max Activating Token Index: 420

Text #1

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0367. Min Act: -0.0001

Data Index: 1042873 (C4 (Web Text))

Max Activating Token Index: 135

Text #2

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0331. Min Act: -0.0002

Data Index: 88798 (C4 (Web Text))

Max Activating Token Index: 104

Text #3

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0330. Min Act: -0.0001

Data Index: 950095 (C4 (Web Text))

Max Activating Token Index: 259

Text #4

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0331. Min Act: -0.0002

Data Index: 527996 (C4 (Web Text))

Max Activating Token Index: 802

Text #5

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0322. Min Act: -0.0002

Data Index: 320546 (C4 (Web Text))

Max Activating Token Index: 423

Text #6

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0327. Min Act: -0.0002

Data Index: 875308 (C4 (Web Text))

Max Activating Token Index: 171

Text #7

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0319. Min Act: -0.0002

Data Index: 807502 (C4 (Web Text))

Max Activating Token Index: 584

Text #8

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0318. Min Act: -0.0001

Data Index: 696899 (C4 (Web Text))

Max Activating Token Index: 577

Text #9

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0308. Min Act: -0.0002

Data Index: 358395 (C4 (Web Text))

Max Activating Token Index: 637

Text #10

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0322. Min Act: -0.0001

Data Index: 718477 (C4 (Web Text))

Max Activating Token Index: 50

Text #11

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0295. Min Act: -0.0001

Data Index: 657658 (C4 (Web Text))

Max Activating Token Index: 826

Text #12

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0313. Min Act: -0.0001

Data Index: 839575 (C4 (Web Text))

Max Activating Token Index: 807

Text #13

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0302. Min Act: -0.0001

Data Index: 151499 (C4 (Web Text))

Max Activating Token Index: 365

Text #14

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0302. Min Act: -0.0002

Data Index: 1287131 (C4 (Web Text))

Max Activating Token Index: 360

Text #15

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0293. Min Act: -0.0002

Data Index: 420165 (C4 (Web Text))

Max Activating Token Index: 106

Text #16

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0298. Min Act: -0.0002

Data Index: 1047168 (C4 (Web Text))

Max Activating Token Index: 648

Text #17

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0288. Min Act: -0.0002

Data Index: 935870 (C4 (Web Text))

Max Activating Token Index: 509

Text #18

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0289. Min Act: -0.0002

Data Index: 142209 (C4 (Web Text))

Max Activating Token Index: 845

Text #19

Max Range: 0.0486. Min Range: -0.0486

Max Act: 0.0291. Min Act: -0.0002

Data Index: 62918 (C4 (Web Text))

Max Activating Token Index: 136