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

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

Neuron 3221 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0501. Min Act: -0.0000

Data Index: 1021458 (C4 (Web Text))

Max Activating Token Index: 344

Text #1

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0456. Min Act: -0.0000

Data Index: 809886 (C4 (Web Text))

Max Activating Token Index: 841

Text #2

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0353. Min Act: -0.0000

Data Index: 113656 (C4 (Web Text))

Max Activating Token Index: 134

Text #3

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0385. Min Act: -0.0000

Data Index: 41800 (C4 (Web Text))

Max Activating Token Index: 942

Text #4

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0347. Min Act: -0.0000

Data Index: 1545950 (Python Code)

Max Activating Token Index: 466

Text #5

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0365. Min Act: -0.0000

Data Index: 992729 (C4 (Web Text))

Max Activating Token Index: 384

Text #6

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0364. Min Act: -0.0000

Data Index: 99865 (C4 (Web Text))

Max Activating Token Index: 605

Text #7

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0332. Min Act: -0.0000

Data Index: 366672 (C4 (Web Text))

Max Activating Token Index: 407

Text #8

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0360. Min Act: -0.0000

Data Index: 589834 (C4 (Web Text))

Max Activating Token Index: 619

Text #9

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0345. Min Act: -0.0000

Data Index: 384348 (C4 (Web Text))

Max Activating Token Index: 696

Text #10

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0322. Min Act: -0.0000

Data Index: 9407 (C4 (Web Text))

Max Activating Token Index: 658

Text #11

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0327. Min Act: -0.0000

Data Index: 1458094 (Python Code)

Max Activating Token Index: 882

Text #12

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0311. Min Act: -0.0000

Data Index: 927595 (C4 (Web Text))

Max Activating Token Index: 922

Text #13

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0324. Min Act: -0.0000

Data Index: 455248 (C4 (Web Text))

Max Activating Token Index: 1016

Text #14

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0312. Min Act: -0.0000

Data Index: 1007631 (C4 (Web Text))

Max Activating Token Index: 96

Text #15

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0321. Min Act: -0.0000

Data Index: 212452 (C4 (Web Text))

Max Activating Token Index: 728

Text #16

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0271. Min Act: -0.0000

Data Index: 485716 (C4 (Web Text))

Max Activating Token Index: 57

Text #17

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0321. Min Act: -0.0000

Data Index: 486359 (C4 (Web Text))

Max Activating Token Index: 681

Text #18

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0315. Min Act: -0.0000

Data Index: 751106 (C4 (Web Text))

Max Activating Token Index: 727

Text #19

Max Range: 0.0501. Min Range: -0.0501

Max Act: 0.0308. Min Act: -0.0000

Data Index: 1476379 (Python Code)

Max Activating Token Index: 338