Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 128 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.6362. Min Act: -0.0001

Data Index: 402792 (C4 (Web Text))

Max Activating Token Index: 518

Text #1

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.5439. Min Act: -0.0001

Data Index: 485988 (C4 (Web Text))

Max Activating Token Index: 874

Text #2

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.4918. Min Act: -0.0001

Data Index: 218519 (C4 (Web Text))

Max Activating Token Index: 737

Text #3

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.5231. Min Act: -0.0001

Data Index: 1012971 (C4 (Web Text))

Max Activating Token Index: 733

Text #4

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.6376. Min Act: -0.0001

Data Index: 600137 (C4 (Web Text))

Max Activating Token Index: 604

Text #5

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.5668. Min Act: -0.0001

Data Index: 595161 (C4 (Web Text))

Max Activating Token Index: 904

Text #6

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.4053. Min Act: -0.0001

Data Index: 619058 (C4 (Web Text))

Max Activating Token Index: 185

Text #7

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.4313. Min Act: -0.0001

Data Index: 446636 (C4 (Web Text))

Max Activating Token Index: 997

Text #8

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3850. Min Act: -0.0001

Data Index: 701923 (C4 (Web Text))

Max Activating Token Index: 383

Text #9

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3842. Min Act: -0.0001

Data Index: 423732 (C4 (Web Text))

Max Activating Token Index: 38

Text #10

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3428. Min Act: -0.0001

Data Index: 231506 (C4 (Web Text))

Max Activating Token Index: 295

Text #11

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3966. Min Act: -0.0001

Data Index: 921022 (C4 (Web Text))

Max Activating Token Index: 365

Text #12

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3379. Min Act: -0.0001

Data Index: 1650454 (Python Code)

Max Activating Token Index: 525

Text #13

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3672. Min Act: -0.0001

Data Index: 357762 (C4 (Web Text))

Max Activating Token Index: 802

Text #14

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3584. Min Act: -0.0001

Data Index: 532861 (C4 (Web Text))

Max Activating Token Index: 788

Text #15

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3680. Min Act: -0.0001

Data Index: 1486131 (Python Code)

Max Activating Token Index: 116

Text #16

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3716. Min Act: -0.0001

Data Index: 541170 (C4 (Web Text))

Max Activating Token Index: 589

Text #17

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3545. Min Act: -0.0001

Data Index: 1407898 (Python Code)

Max Activating Token Index: 690

Text #18

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3356. Min Act: -0.0001

Data Index: 1464284 (Python Code)

Max Activating Token Index: 364

Text #19

Max Range: 0.6376. Min Range: -0.6376

Max Act: 0.3469. Min Act: -0.0001

Data Index: 337509 (C4 (Web Text))

Max Activating Token Index: 918