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

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

Neuron 4922 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.8440. Min Act: -0.0000

Data Index: 1411691 (Python Code)

Max Activating Token Index: 204

Text #1

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.3944. Min Act: -0.0000

Data Index: 1435298 (Python Code)

Max Activating Token Index: 368

Text #2

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.2491. Min Act: -0.0000

Data Index: 1628419 (Python Code)

Max Activating Token Index: 425

Text #3

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.3432. Min Act: -0.0000

Data Index: 1609367 (Python Code)

Max Activating Token Index: 423

Text #4

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.3331. Min Act: -0.0000

Data Index: 390656 (C4 (Web Text))

Max Activating Token Index: 514

Text #5

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.0487. Min Act: -0.0000

Data Index: 1264263 (C4 (Web Text))

Max Activating Token Index: 231

Text #6

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.0411. Min Act: -0.0000

Data Index: 892765 (C4 (Web Text))

Max Activating Token Index: 693

Text #7

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.1307. Min Act: -0.0000

Data Index: 944616 (C4 (Web Text))

Max Activating Token Index: 338

Text #8

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.1351. Min Act: -0.0000

Data Index: 746811 (C4 (Web Text))

Max Activating Token Index: 528

Text #9

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.1291. Min Act: -0.0000

Data Index: 2481 (C4 (Web Text))

Max Activating Token Index: 277

Text #10

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.1078. Min Act: -0.0000

Data Index: 1103734 (C4 (Web Text))

Max Activating Token Index: 869

Text #11

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.0461. Min Act: -0.0000

Data Index: 1430619 (Python Code)

Max Activating Token Index: 527

Text #12

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.1368. Min Act: -0.0000

Data Index: 518489 (C4 (Web Text))

Max Activating Token Index: 143

Text #13

Max Range: 2.8440. Min Range: -2.8440

Max Act: 1.9962. Min Act: -0.0000

Data Index: 316510 (C4 (Web Text))

Max Activating Token Index: 973

Text #14

Max Range: 2.8440. Min Range: -2.8440

Max Act: 2.0915. Min Act: -0.0000

Data Index: 1212121 (C4 (Web Text))

Max Activating Token Index: 613

Text #15

Max Range: 2.8440. Min Range: -2.8440

Max Act: 1.9768. Min Act: -0.0000

Data Index: 1087767 (C4 (Web Text))

Max Activating Token Index: 609

Text #16

Max Range: 2.8440. Min Range: -2.8440

Max Act: 1.9737. Min Act: -0.0000

Data Index: 860162 (C4 (Web Text))

Max Activating Token Index: 631

Text #17

Max Range: 2.8440. Min Range: -2.8440

Max Act: 1.9672. Min Act: -0.0000

Data Index: 340380 (C4 (Web Text))

Max Activating Token Index: 544

Text #18

Max Range: 2.8440. Min Range: -2.8440

Max Act: 1.9397. Min Act: -0.0000

Data Index: 1096100 (C4 (Web Text))

Max Activating Token Index: 557

Text #19

Max Range: 2.8440. Min Range: -2.8440

Max Act: 1.9379. Min Act: -0.0000

Data Index: 814495 (C4 (Web Text))

Max Activating Token Index: 634