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

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

Neuron 2475 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1416. Min Range: -2.1416

Max Act: 2.1416. Min Act: -0.0001

Data Index: 1468683 (Python Code)

Max Activating Token Index: 968

Text #1

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.8950. Min Act: -0.0001

Data Index: 1207636 (C4 (Web Text))

Max Activating Token Index: 529

Text #2

Max Range: 2.1416. Min Range: -2.1416

Max Act: 2.0754. Min Act: -0.0000

Data Index: 1649687 (Python Code)

Max Activating Token Index: 1004

Text #3

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.8356. Min Act: -0.0000

Data Index: 301493 (C4 (Web Text))

Max Activating Token Index: 268

Text #4

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.9157. Min Act: -0.0000

Data Index: 497249 (C4 (Web Text))

Max Activating Token Index: 716

Text #5

Max Range: 2.1416. Min Range: -2.1416

Max Act: 2.0069. Min Act: -0.0000

Data Index: 567372 (C4 (Web Text))

Max Activating Token Index: 722

Text #6

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.9104. Min Act: -0.0000

Data Index: 310968 (C4 (Web Text))

Max Activating Token Index: 669

Text #7

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.7732. Min Act: -0.0000

Data Index: 881252 (C4 (Web Text))

Max Activating Token Index: 511

Text #8

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.8912. Min Act: -0.0001

Data Index: 931990 (C4 (Web Text))

Max Activating Token Index: 473

Text #9

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.8815. Min Act: -0.0000

Data Index: 1157317 (C4 (Web Text))

Max Activating Token Index: 1000

Text #10

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.8248. Min Act: -0.0001

Data Index: 1009730 (C4 (Web Text))

Max Activating Token Index: 466

Text #11

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.8114. Min Act: -0.0000

Data Index: 402515 (C4 (Web Text))

Max Activating Token Index: 364

Text #12

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.8724. Min Act: -0.0000

Data Index: 737408 (C4 (Web Text))

Max Activating Token Index: 977

Text #13

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.7789. Min Act: -0.0000

Data Index: 348547 (C4 (Web Text))

Max Activating Token Index: 943

Text #14

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.7271. Min Act: -0.0000

Data Index: 533156 (C4 (Web Text))

Max Activating Token Index: 995

Text #15

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.8231. Min Act: -0.0001

Data Index: 1400623 (Python Code)

Max Activating Token Index: 799

Text #16

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.7788. Min Act: -0.0001

Data Index: 75510 (C4 (Web Text))

Max Activating Token Index: 478

Text #17

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.7630. Min Act: -0.0001

Data Index: 1449286 (Python Code)

Max Activating Token Index: 518

Text #18

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.6232. Min Act: -0.0001

Data Index: 1337411 (C4 (Web Text))

Max Activating Token Index: 374

Text #19

Max Range: 2.1416. Min Range: -2.1416

Max Act: 1.8507. Min Act: -0.0000

Data Index: 466328 (C4 (Web Text))

Max Activating Token Index: 806