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

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

Neuron 684 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.4003. Min Act: -0.0000

Data Index: 1606940 (Python Code)

Max Activating Token Index: 450

Text #1

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.3740. Min Act: -0.0000

Data Index: 452831 (C4 (Web Text))

Max Activating Token Index: 711

Text #2

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.2525. Min Act: -0.0000

Data Index: 840900 (C4 (Web Text))

Max Activating Token Index: 501

Text #3

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.1502. Min Act: -0.0000

Data Index: 960973 (C4 (Web Text))

Max Activating Token Index: 292

Text #4

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.2468. Min Act: -0.0000

Data Index: 533375 (C4 (Web Text))

Max Activating Token Index: 412

Text #5

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.2363. Min Act: -0.0000

Data Index: 634911 (C4 (Web Text))

Max Activating Token Index: 562

Text #6

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.2201. Min Act: -0.0000

Data Index: 1443749 (Python Code)

Max Activating Token Index: 566

Text #7

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.1588. Min Act: -0.0000

Data Index: 877554 (C4 (Web Text))

Max Activating Token Index: 934

Text #8

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.2039. Min Act: -0.0000

Data Index: 242775 (C4 (Web Text))

Max Activating Token Index: 662

Text #9

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.1169. Min Act: -0.0000

Data Index: 747614 (C4 (Web Text))

Max Activating Token Index: 705

Text #10

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.0776. Min Act: -0.0000

Data Index: 1647460 (Python Code)

Max Activating Token Index: 168

Text #11

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.0820. Min Act: -0.0000

Data Index: 69046 (C4 (Web Text))

Max Activating Token Index: 173

Text #12

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.1025. Min Act: -0.0000

Data Index: 1551446 (Python Code)

Max Activating Token Index: 943

Text #13

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.0022. Min Act: -0.0000

Data Index: 114648 (C4 (Web Text))

Max Activating Token Index: 638

Text #14

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.1232. Min Act: -0.0000

Data Index: 1634537 (Python Code)

Max Activating Token Index: 774

Text #15

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.0530. Min Act: -0.0000

Data Index: 1457188 (Python Code)

Max Activating Token Index: 639

Text #16

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.1361. Min Act: -0.0000

Data Index: 31430 (C4 (Web Text))

Max Activating Token Index: 949

Text #17

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.0050. Min Act: -0.0000

Data Index: 287769 (C4 (Web Text))

Max Activating Token Index: 257

Text #18

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.0076. Min Act: -0.0000

Data Index: 207787 (C4 (Web Text))

Max Activating Token Index: 385

Text #19

Max Range: 1.4003. Min Range: -1.4003

Max Act: 1.1296. Min Act: -0.0000

Data Index: 336534 (C4 (Web Text))

Max Activating Token Index: 46