Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5117 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0484. Min Act: -0.0001

Data Index: 545250 (C4 (Web Text))

Max Activating Token Index: 828

Text #1

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0433. Min Act: -0.0001

Data Index: 1270741 (C4 (Web Text))

Max Activating Token Index: 336

Text #2

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0380. Min Act: -0.0001

Data Index: 272214 (C4 (Web Text))

Max Activating Token Index: 403

Text #3

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0393. Min Act: -0.0001

Data Index: 468677 (C4 (Web Text))

Max Activating Token Index: 253

Text #4

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0397. Min Act: -0.0001

Data Index: 766791 (C4 (Web Text))

Max Activating Token Index: 509

Text #5

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0394. Min Act: -0.0001

Data Index: 720743 (C4 (Web Text))

Max Activating Token Index: 824

Text #6

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0352. Min Act: -0.0001

Data Index: 654805 (C4 (Web Text))

Max Activating Token Index: 124

Text #7

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0335. Min Act: -0.0001

Data Index: 737688 (C4 (Web Text))

Max Activating Token Index: 440

Text #8

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0298. Min Act: -0.0001

Data Index: 2873 (C4 (Web Text))

Max Activating Token Index: 141

Text #9

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0325. Min Act: -0.0001

Data Index: 286060 (C4 (Web Text))

Max Activating Token Index: 162

Text #10

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0317. Min Act: -0.0001

Data Index: 709818 (C4 (Web Text))

Max Activating Token Index: 482

Text #11

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0307. Min Act: -0.0001

Data Index: 1277654 (C4 (Web Text))

Max Activating Token Index: 496

Text #12

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0328. Min Act: -0.0001

Data Index: 422428 (C4 (Web Text))

Max Activating Token Index: 610

Text #13

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0325. Min Act: -0.0001

Data Index: 695794 (C4 (Web Text))

Max Activating Token Index: 796

Text #14

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0299. Min Act: -0.0001

Data Index: 299116 (C4 (Web Text))

Max Activating Token Index: 160

Text #15

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0336. Min Act: -0.0001

Data Index: 1454583 (Python Code)

Max Activating Token Index: 944

Text #16

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0271. Min Act: -0.0001

Data Index: 61454 (C4 (Web Text))

Max Activating Token Index: 869

Text #17

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0342. Min Act: -0.0001

Data Index: 1135638 (C4 (Web Text))

Max Activating Token Index: 956

Text #18

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0280. Min Act: -0.0001

Data Index: 157994 (C4 (Web Text))

Max Activating Token Index: 317

Text #19

Max Range: 0.0484. Min Range: -0.0484

Max Act: 0.0289. Min Act: -0.0001

Data Index: 1145273 (C4 (Web Text))

Max Activating Token Index: 446