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

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

Neuron 1249 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.6494. Min Act: -0.0000

Data Index: 1443144 (Python Code)

Max Activating Token Index: 933

Text #1

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.5503. Min Act: -0.0000

Data Index: 1435047 (Python Code)

Max Activating Token Index: 801

Text #2

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.5588. Min Act: -0.0000

Data Index: 1249925 (C4 (Web Text))

Max Activating Token Index: 962

Text #3

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.5434. Min Act: -0.0000

Data Index: 398741 (C4 (Web Text))

Max Activating Token Index: 604

Text #4

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.5256. Min Act: -0.0000

Data Index: 48679 (C4 (Web Text))

Max Activating Token Index: 608

Text #5

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4943. Min Act: -0.0000

Data Index: 1300262 (C4 (Web Text))

Max Activating Token Index: 752

Text #6

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4731. Min Act: -0.0000

Data Index: 222523 (C4 (Web Text))

Max Activating Token Index: 460

Text #7

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4722. Min Act: -0.0000

Data Index: 1551858 (Python Code)

Max Activating Token Index: 467

Text #8

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4394. Min Act: -0.0000

Data Index: 1653806 (Python Code)

Max Activating Token Index: 228

Text #9

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4901. Min Act: -0.0000

Data Index: 667296 (C4 (Web Text))

Max Activating Token Index: 953

Text #10

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4543. Min Act: -0.0000

Data Index: 471962 (C4 (Web Text))

Max Activating Token Index: 870

Text #11

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4331. Min Act: -0.0000

Data Index: 1547630 (Python Code)

Max Activating Token Index: 22

Text #12

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4182. Min Act: -0.0000

Data Index: 772446 (C4 (Web Text))

Max Activating Token Index: 576

Text #13

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4299. Min Act: -0.0000

Data Index: 335877 (C4 (Web Text))

Max Activating Token Index: 694

Text #14

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4260. Min Act: -0.0000

Data Index: 1201544 (C4 (Web Text))

Max Activating Token Index: 468

Text #15

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4272. Min Act: -0.0000

Data Index: 1033924 (C4 (Web Text))

Max Activating Token Index: 509

Text #16

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4247. Min Act: -0.0000

Data Index: 1146563 (C4 (Web Text))

Max Activating Token Index: 588

Text #17

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.4513. Min Act: -0.0000

Data Index: 1649248 (Python Code)

Max Activating Token Index: 412

Text #18

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.3566. Min Act: -0.0000

Data Index: 1447900 (Python Code)

Max Activating Token Index: 366

Text #19

Max Range: 0.6494. Min Range: -0.6494

Max Act: 0.3871. Min Act: -0.0000

Data Index: 745234 (C4 (Web Text))

Max Activating Token Index: 985