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

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

Neuron 3926 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.9482. Min Act: -0.0001

Data Index: 1634091 (Python Code)

Max Activating Token Index: 878

Text #1

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.8651. Min Act: -0.0001

Data Index: 1384131 (Python Code)

Max Activating Token Index: 420

Text #2

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.7844. Min Act: -0.0001

Data Index: 1473449 (Python Code)

Max Activating Token Index: 792

Text #3

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.7777. Min Act: -0.0001

Data Index: 1002715 (C4 (Web Text))

Max Activating Token Index: 1010

Text #4

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.7628. Min Act: -0.0001

Data Index: 1640518 (Python Code)

Max Activating Token Index: 475

Text #5

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.7581. Min Act: -0.0001

Data Index: 1143092 (C4 (Web Text))

Max Activating Token Index: 840

Text #6

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.7564. Min Act: -0.0001

Data Index: 1481057 (Python Code)

Max Activating Token Index: 15

Text #7

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.8072. Min Act: -0.0001

Data Index: 518642 (C4 (Web Text))

Max Activating Token Index: 360

Text #8

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.7415. Min Act: -0.0001

Data Index: 107956 (C4 (Web Text))

Max Activating Token Index: 620

Text #9

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.7222. Min Act: -0.0001

Data Index: 1410689 (Python Code)

Max Activating Token Index: 1007

Text #10

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6780. Min Act: -0.0001

Data Index: 625740 (C4 (Web Text))

Max Activating Token Index: 522

Text #11

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6848. Min Act: -0.0001

Data Index: 1513818 (Python Code)

Max Activating Token Index: 472

Text #12

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6949. Min Act: -0.0001

Data Index: 157766 (C4 (Web Text))

Max Activating Token Index: 499

Text #13

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6773. Min Act: -0.0001

Data Index: 1624489 (Python Code)

Max Activating Token Index: 81

Text #14

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6861. Min Act: -0.0001

Data Index: 1490533 (Python Code)

Max Activating Token Index: 1021

Text #15

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6877. Min Act: -0.0001

Data Index: 1493677 (Python Code)

Max Activating Token Index: 479

Text #16

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6759. Min Act: -0.0001

Data Index: 1396152 (Python Code)

Max Activating Token Index: 284

Text #17

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6596. Min Act: -0.0001

Data Index: 1426900 (Python Code)

Max Activating Token Index: 326

Text #18

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6664. Min Act: -0.0001

Data Index: 1491297 (Python Code)

Max Activating Token Index: 84

Text #19

Max Range: 0.9482. Min Range: -0.9482

Max Act: 0.6218. Min Act: -0.0001

Data Index: 1061191 (C4 (Web Text))

Max Activating Token Index: 36