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

Dataset: The Pile

Neuron 1344 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 9.1739. Min Range: -9.1739

Max Act: 9.0654. Min Act: -0.0000

Data Index: 260181 (The Pile)

Max Activating Token Index: 148

Text #1

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.5548. Min Act: -0.0000

Data Index: 1571009 (The Pile)

Max Activating Token Index: 580

Text #2

Max Range: 9.1739. Min Range: -9.1739

Max Act: 9.1739. Min Act: -0.0000

Data Index: 139917 (The Pile)

Max Activating Token Index: 815

Text #3

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.9661. Min Act: -0.0000

Data Index: 1268734 (The Pile)

Max Activating Token Index: 773

Text #4

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.8166. Min Act: -0.0000

Data Index: 914852 (The Pile)

Max Activating Token Index: 480

Text #5

Max Range: 9.1739. Min Range: -9.1739

Max Act: 9.0729. Min Act: -0.0000

Data Index: 621172 (The Pile)

Max Activating Token Index: 158

Text #6

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.8574. Min Act: -0.0000

Data Index: 473799 (The Pile)

Max Activating Token Index: 945

Text #7

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.7015. Min Act: -0.0000

Data Index: 1015025 (The Pile)

Max Activating Token Index: 262

Text #8

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.6853. Min Act: -0.0000

Data Index: 246760 (The Pile)

Max Activating Token Index: 815

Text #9

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.7637. Min Act: -0.0000

Data Index: 653251 (The Pile)

Max Activating Token Index: 710

Text #10

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.7280. Min Act: -0.0000

Data Index: 1100529 (The Pile)

Max Activating Token Index: 596

Text #11

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.5580. Min Act: -0.0000

Data Index: 907111 (The Pile)

Max Activating Token Index: 436

Text #12

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.2211. Min Act: -0.0000

Data Index: 649888 (The Pile)

Max Activating Token Index: 501

Text #13

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.6818. Min Act: -0.0000

Data Index: 389436 (The Pile)

Max Activating Token Index: 751

Text #14

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.4663. Min Act: -0.0000

Data Index: 764190 (The Pile)

Max Activating Token Index: 656

Text #15

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.4656. Min Act: -0.0000

Data Index: 554510 (The Pile)

Max Activating Token Index: 787

Text #16

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.4218. Min Act: -0.0000

Data Index: 658562 (The Pile)

Max Activating Token Index: 564

Text #17

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.2652. Min Act: -0.0000

Data Index: 1260023 (The Pile)

Max Activating Token Index: 757

Text #18

Max Range: 9.1739. Min Range: -9.1739

Max Act: 8.5831. Min Act: -0.0000

Data Index: 223145 (The Pile)

Max Activating Token Index: 652

Text #19

Max Range: 9.1739. Min Range: -9.1739

Max Act: 7.7942. Min Act: -0.0000

Data Index: 68362 (The Pile)

Max Activating Token Index: 952