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

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

Neuron 4558 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: 3.4578. Min Range: -3.4578

Max Act: 3.4492. Min Act: -0.0000

Data Index: 1448122 (Python Code)

Max Activating Token Index: 570

Text #1

Max Range: 3.4578. Min Range: -3.4578

Max Act: 3.1426. Min Act: -0.0000

Data Index: 447328 (C4 (Web Text))

Max Activating Token Index: 627

Text #2

Max Range: 3.4578. Min Range: -3.4578

Max Act: 3.4578. Min Act: -0.0000

Data Index: 1049056 (C4 (Web Text))

Max Activating Token Index: 814

Text #3

Max Range: 3.4578. Min Range: -3.4578

Max Act: 3.1126. Min Act: -0.0000

Data Index: 1027819 (C4 (Web Text))

Max Activating Token Index: 883

Text #4

Max Range: 3.4578. Min Range: -3.4578

Max Act: 3.1915. Min Act: -0.0000

Data Index: 1443871 (Python Code)

Max Activating Token Index: 481

Text #5

Max Range: 3.4578. Min Range: -3.4578

Max Act: 3.0599. Min Act: -0.0000

Data Index: 1437946 (Python Code)

Max Activating Token Index: 293

Text #6

Max Range: 3.4578. Min Range: -3.4578

Max Act: 3.2931. Min Act: -0.0000

Data Index: 29011 (C4 (Web Text))

Max Activating Token Index: 435

Text #7

Max Range: 3.4578. Min Range: -3.4578

Max Act: 3.0354. Min Act: -0.0000

Data Index: 1585246 (Python Code)

Max Activating Token Index: 73

Text #8

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.7628. Min Act: -0.0000

Data Index: 125921 (C4 (Web Text))

Max Activating Token Index: 928

Text #9

Max Range: 3.4578. Min Range: -3.4578

Max Act: 3.0378. Min Act: -0.0000

Data Index: 865120 (C4 (Web Text))

Max Activating Token Index: 883

Text #10

Max Range: 3.4578. Min Range: -3.4578

Max Act: 3.0358. Min Act: -0.0000

Data Index: 1130470 (C4 (Web Text))

Max Activating Token Index: 947

Text #11

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.6264. Min Act: -0.0000

Data Index: 53206 (C4 (Web Text))

Max Activating Token Index: 138

Text #12

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.8889. Min Act: -0.0000

Data Index: 1520038 (Python Code)

Max Activating Token Index: 916

Text #13

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.9048. Min Act: -0.0000

Data Index: 1583086 (Python Code)

Max Activating Token Index: 931

Text #14

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.7574. Min Act: -0.0000

Data Index: 926455 (C4 (Web Text))

Max Activating Token Index: 383

Text #15

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.9840. Min Act: -0.0000

Data Index: 1241912 (C4 (Web Text))

Max Activating Token Index: 371

Text #16

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.9784. Min Act: -0.0000

Data Index: 1657042 (Python Code)

Max Activating Token Index: 966

Text #17

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.9648. Min Act: -0.0000

Data Index: 1548406 (Python Code)

Max Activating Token Index: 499

Text #18

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.7859. Min Act: -0.0000

Data Index: 1634438 (Python Code)

Max Activating Token Index: 702

Text #19

Max Range: 3.4578. Min Range: -3.4578

Max Act: 2.6579. Min Act: -0.0000

Data Index: 2778 (C4 (Web Text))

Max Activating Token Index: 679