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

Dataset: The Pile

Neuron 3615 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8867. Min Range: -3.8867

Max Act: 3.8290. Min Act: -0.0000

Data Index: 245823 (The Pile)

Max Activating Token Index: 794

Text #1

Max Range: 3.8867. Min Range: -3.8867

Max Act: 3.8867. Min Act: -0.0000

Data Index: 55829 (The Pile)

Max Activating Token Index: 752

Text #2

Max Range: 3.8867. Min Range: -3.8867

Max Act: 3.0356. Min Act: -0.0000

Data Index: 125274 (The Pile)

Max Activating Token Index: 929

Text #3

Max Range: 3.8867. Min Range: -3.8867

Max Act: 3.0910. Min Act: -0.0000

Data Index: 42868 (The Pile)

Max Activating Token Index: 88

Text #4

Max Range: 3.8867. Min Range: -3.8867

Max Act: 3.0450. Min Act: -0.0000

Data Index: 1994830 (The Pile)

Max Activating Token Index: 876

Text #5

Max Range: 3.8867. Min Range: -3.8867

Max Act: 3.1887. Min Act: -0.0000

Data Index: 236171 (The Pile)

Max Activating Token Index: 470

Text #6

Max Range: 3.8867. Min Range: -3.8867

Max Act: 3.0414. Min Act: -0.0000

Data Index: 370846 (The Pile)

Max Activating Token Index: 454

Text #7

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.9801. Min Act: -0.0000

Data Index: 326251 (The Pile)

Max Activating Token Index: 523

Text #8

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.8989. Min Act: -0.0000

Data Index: 1607691 (The Pile)

Max Activating Token Index: 499

Text #9

Max Range: 3.8867. Min Range: -3.8867

Max Act: 3.0391. Min Act: -0.0000

Data Index: 549626 (The Pile)

Max Activating Token Index: 379

Text #10

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.7417. Min Act: -0.0000

Data Index: 1562458 (The Pile)

Max Activating Token Index: 405

Text #11

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.7336. Min Act: -0.0000

Data Index: 1016093 (The Pile)

Max Activating Token Index: 187

Text #12

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.8679. Min Act: -0.0000

Data Index: 1012923 (The Pile)

Max Activating Token Index: 846

Text #13

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.8566. Min Act: -0.0000

Data Index: 1591097 (The Pile)

Max Activating Token Index: 984

Text #14

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.8283. Min Act: -0.0000

Data Index: 606854 (The Pile)

Max Activating Token Index: 570

Text #15

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.8396. Min Act: -0.0000

Data Index: 348387 (The Pile)

Max Activating Token Index: 120

Text #16

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.9640. Min Act: -0.0000

Data Index: 209565 (The Pile)

Max Activating Token Index: 876

Text #17

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.8414. Min Act: -0.0000

Data Index: 344681 (The Pile)

Max Activating Token Index: 193

Text #18

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.6871. Min Act: -0.0000

Data Index: 1076339 (The Pile)

Max Activating Token Index: 58

Text #19

Max Range: 3.8867. Min Range: -3.8867

Max Act: 2.8095. Min Act: -0.0000

Data Index: 1218285 (The Pile)

Max Activating Token Index: 295