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

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

Neuron 2017 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0621. Min Range: -1.0621

Max Act: 1.0621. Min Act: -0.0000

Data Index: 762341 (C4 (Web Text))

Max Activating Token Index: 64

Text #1

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.8827. Min Act: -0.0000

Data Index: 639101 (C4 (Web Text))

Max Activating Token Index: 839

Text #2

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.8727. Min Act: -0.0000

Data Index: 987051 (C4 (Web Text))

Max Activating Token Index: 836

Text #3

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.8681. Min Act: -0.0000

Data Index: 1209580 (C4 (Web Text))

Max Activating Token Index: 226

Text #4

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.8612. Min Act: -0.0000

Data Index: 740388 (C4 (Web Text))

Max Activating Token Index: 794

Text #5

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.8317. Min Act: -0.0000

Data Index: 361673 (C4 (Web Text))

Max Activating Token Index: 466

Text #6

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.8013. Min Act: -0.0000

Data Index: 473438 (C4 (Web Text))

Max Activating Token Index: 414

Text #7

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7903. Min Act: -0.0000

Data Index: 357204 (C4 (Web Text))

Max Activating Token Index: 646

Text #8

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.8387. Min Act: -0.0000

Data Index: 873686 (C4 (Web Text))

Max Activating Token Index: 963

Text #9

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7871. Min Act: -0.0000

Data Index: 1296803 (C4 (Web Text))

Max Activating Token Index: 234

Text #10

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7842. Min Act: -0.0000

Data Index: 225818 (C4 (Web Text))

Max Activating Token Index: 836

Text #11

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7388. Min Act: -0.0000

Data Index: 382518 (C4 (Web Text))

Max Activating Token Index: 895

Text #12

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7626. Min Act: -0.0000

Data Index: 776369 (C4 (Web Text))

Max Activating Token Index: 307

Text #13

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.8129. Min Act: -0.0000

Data Index: 897182 (C4 (Web Text))

Max Activating Token Index: 96

Text #14

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7601. Min Act: -0.0000

Data Index: 1110621 (C4 (Web Text))

Max Activating Token Index: 203

Text #15

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7351. Min Act: -0.0000

Data Index: 676511 (C4 (Web Text))

Max Activating Token Index: 652

Text #16

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7344. Min Act: -0.0000

Data Index: 696875 (C4 (Web Text))

Max Activating Token Index: 584

Text #17

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7450. Min Act: -0.0000

Data Index: 268208 (C4 (Web Text))

Max Activating Token Index: 181

Text #18

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7533. Min Act: -0.0000

Data Index: 213794 (C4 (Web Text))

Max Activating Token Index: 946

Text #19

Max Range: 1.0621. Min Range: -1.0621

Max Act: 0.7265. Min Act: -0.0000

Data Index: 365180 (C4 (Web Text))

Max Activating Token Index: 482