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

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

Neuron 70 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.4273. Min Act: -0.0002

Data Index: 1295035 (C4 (Web Text))

Max Activating Token Index: 609

Text #1

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.3305. Min Act: -0.0001

Data Index: 1308485 (C4 (Web Text))

Max Activating Token Index: 1007

Text #2

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.3209. Min Act: -0.0001

Data Index: 1483221 (Python Code)

Max Activating Token Index: 516

Text #3

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.3958. Min Act: -0.0001

Data Index: 1394659 (Python Code)

Max Activating Token Index: 49

Text #4

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.3541. Min Act: -0.0002

Data Index: 1063659 (C4 (Web Text))

Max Activating Token Index: 197

Text #5

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.2931. Min Act: -0.0002

Data Index: 707354 (C4 (Web Text))

Max Activating Token Index: 320

Text #6

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.2061. Min Act: -0.0001

Data Index: 802661 (C4 (Web Text))

Max Activating Token Index: 524

Text #7

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.3581. Min Act: -0.0002

Data Index: 912197 (C4 (Web Text))

Max Activating Token Index: 618

Text #8

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.1873. Min Act: -0.0002

Data Index: 705673 (C4 (Web Text))

Max Activating Token Index: 782

Text #9

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.2636. Min Act: -0.0002

Data Index: 1131733 (C4 (Web Text))

Max Activating Token Index: 454

Text #10

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.1902. Min Act: -0.0001

Data Index: 1120787 (C4 (Web Text))

Max Activating Token Index: 184

Text #11

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.3040. Min Act: -0.0002

Data Index: 591040 (C4 (Web Text))

Max Activating Token Index: 131

Text #12

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.1584. Min Act: -0.0002

Data Index: 460232 (C4 (Web Text))

Max Activating Token Index: 494

Text #13

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.1481. Min Act: -0.0001

Data Index: 1616699 (Python Code)

Max Activating Token Index: 211

Text #14

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.1493. Min Act: -0.0001

Data Index: 1614315 (Python Code)

Max Activating Token Index: 914

Text #15

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.1078. Min Act: -0.0001

Data Index: 1384046 (Python Code)

Max Activating Token Index: 232

Text #16

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.1585. Min Act: -0.0001

Data Index: 146108 (C4 (Web Text))

Max Activating Token Index: 855

Text #17

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.1716. Min Act: -0.0002

Data Index: 767908 (C4 (Web Text))

Max Activating Token Index: 909

Text #18

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.1200. Min Act: -0.0002

Data Index: 974943 (C4 (Web Text))

Max Activating Token Index: 708

Text #19

Max Range: 1.4273. Min Range: -1.4273

Max Act: 1.0565. Min Act: -0.0002

Data Index: 402336 (C4 (Web Text))

Max Activating Token Index: 881