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

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

Neuron 471 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0326. Min Act: -0.0005

Data Index: 517965 (C4 (Web Text))

Max Activating Token Index: 222

Text #1

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0342. Min Act: -0.0005

Data Index: 1131246 (C4 (Web Text))

Max Activating Token Index: 380

Text #2

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0320. Min Act: -0.0005

Data Index: 84284 (C4 (Web Text))

Max Activating Token Index: 536

Text #3

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0307. Min Act: -0.0005

Data Index: 1578271 (Python Code)

Max Activating Token Index: 101

Text #4

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0300. Min Act: -0.0005

Data Index: 807583 (C4 (Web Text))

Max Activating Token Index: 451

Text #5

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0310. Min Act: -0.0005

Data Index: 927497 (C4 (Web Text))

Max Activating Token Index: 397

Text #6

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0298. Min Act: -0.0005

Data Index: 1599369 (Python Code)

Max Activating Token Index: 240

Text #7

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0275. Min Act: -0.0005

Data Index: 424633 (C4 (Web Text))

Max Activating Token Index: 460

Text #8

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0292. Min Act: -0.0005

Data Index: 1540714 (Python Code)

Max Activating Token Index: 79

Text #9

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0297. Min Act: -0.0005

Data Index: 1211467 (C4 (Web Text))

Max Activating Token Index: 465

Text #10

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0310. Min Act: -0.0005

Data Index: 1283110 (C4 (Web Text))

Max Activating Token Index: 796

Text #11

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0296. Min Act: -0.0005

Data Index: 762990 (C4 (Web Text))

Max Activating Token Index: 665

Text #12

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0282. Min Act: -0.0005

Data Index: 481028 (C4 (Web Text))

Max Activating Token Index: 468

Text #13

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0272. Min Act: -0.0005

Data Index: 230992 (C4 (Web Text))

Max Activating Token Index: 452

Text #14

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0289. Min Act: -0.0005

Data Index: 1541311 (Python Code)

Max Activating Token Index: 729

Text #15

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0285. Min Act: -0.0005

Data Index: 1364070 (Python Code)

Max Activating Token Index: 89

Text #16

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0289. Min Act: -0.0005

Data Index: 188018 (C4 (Web Text))

Max Activating Token Index: 210

Text #17

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0276. Min Act: -0.0005

Data Index: 7810 (C4 (Web Text))

Max Activating Token Index: 959

Text #18

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0292. Min Act: -0.0005

Data Index: 1552209 (Python Code)

Max Activating Token Index: 644

Text #19

Max Range: 0.0342. Min Range: -0.0342

Max Act: 0.0281. Min Act: -0.0005

Data Index: 247472 (C4 (Web Text))

Max Activating Token Index: 9