Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5525 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 3.4789. Min Range: -3.4789

Max Act: 3.4789. Min Act: -0.0000

Data Index: 80508 (C4 (Web Text))

Max Activating Token Index: 947

Text #1

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.6023. Min Act: -0.0000

Data Index: 819463 (C4 (Web Text))

Max Activating Token Index: 789

Text #2

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.0832. Min Act: -0.0000

Data Index: 495183 (C4 (Web Text))

Max Activating Token Index: 922

Text #3

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.4989. Min Act: -0.0000

Data Index: 472066 (C4 (Web Text))

Max Activating Token Index: 554

Text #4

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.2366. Min Act: -0.0000

Data Index: 1656521 (Python Code)

Max Activating Token Index: 780

Text #5

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.3033. Min Act: -0.0000

Data Index: 1152151 (C4 (Web Text))

Max Activating Token Index: 46

Text #6

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.3888. Min Act: -0.0000

Data Index: 549566 (C4 (Web Text))

Max Activating Token Index: 51

Text #7

Max Range: 3.4789. Min Range: -3.4789

Max Act: 1.8559. Min Act: -0.0000

Data Index: 622473 (C4 (Web Text))

Max Activating Token Index: 781

Text #8

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.3576. Min Act: -0.0000

Data Index: 630973 (C4 (Web Text))

Max Activating Token Index: 238

Text #9

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.2415. Min Act: -0.0000

Data Index: 603778 (C4 (Web Text))

Max Activating Token Index: 182

Text #10

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.1906. Min Act: -0.0000

Data Index: 781910 (C4 (Web Text))

Max Activating Token Index: 72

Text #11

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.1873. Min Act: -0.0000

Data Index: 1637507 (Python Code)

Max Activating Token Index: 214

Text #12

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.0798. Min Act: -0.0000

Data Index: 397567 (C4 (Web Text))

Max Activating Token Index: 122

Text #13

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.0054. Min Act: -0.0000

Data Index: 994299 (C4 (Web Text))

Max Activating Token Index: 303

Text #14

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.0208. Min Act: -0.0000

Data Index: 1154343 (C4 (Web Text))

Max Activating Token Index: 758

Text #15

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.0712. Min Act: -0.0000

Data Index: 716536 (C4 (Web Text))

Max Activating Token Index: 734

Text #16

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.0748. Min Act: -0.0000

Data Index: 997240 (C4 (Web Text))

Max Activating Token Index: 459

Text #17

Max Range: 3.4789. Min Range: -3.4789

Max Act: 1.9444. Min Act: -0.0000

Data Index: 1209807 (C4 (Web Text))

Max Activating Token Index: 684

Text #18

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.1466. Min Act: -0.0000

Data Index: 1626777 (Python Code)

Max Activating Token Index: 959

Text #19

Max Range: 3.4789. Min Range: -3.4789

Max Act: 2.1080. Min Act: -0.0000

Data Index: 784028 (C4 (Web Text))

Max Activating Token Index: 917