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

Dataset: The Pile

Neuron 307 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.7088. Min Act: -0.0000

Data Index: 1658826 (The Pile)

Max Activating Token Index: 886

Text #1

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.6402. Min Act: -0.0000

Data Index: 1079272 (The Pile)

Max Activating Token Index: 968

Text #2

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.6058. Min Act: -0.0000

Data Index: 20525 (The Pile)

Max Activating Token Index: 270

Text #3

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.6563. Min Act: -0.0000

Data Index: 961677 (The Pile)

Max Activating Token Index: 825

Text #4

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.5407. Min Act: -0.0000

Data Index: 814854 (The Pile)

Max Activating Token Index: 381

Text #5

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.4473. Min Act: -0.0000

Data Index: 1028434 (The Pile)

Max Activating Token Index: 146

Text #6

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.4778. Min Act: -0.0000

Data Index: 1157468 (The Pile)

Max Activating Token Index: 721

Text #7

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.4892. Min Act: -0.0000

Data Index: 210838 (The Pile)

Max Activating Token Index: 303

Text #8

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.5745. Min Act: -0.0000

Data Index: 830869 (The Pile)

Max Activating Token Index: 916

Text #9

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.3790. Min Act: -0.0000

Data Index: 718426 (The Pile)

Max Activating Token Index: 469

Text #10

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.3315. Min Act: -0.0000

Data Index: 170675 (The Pile)

Max Activating Token Index: 333

Text #11

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.3836. Min Act: -0.0000

Data Index: 136988 (The Pile)

Max Activating Token Index: 773

Text #12

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.3761. Min Act: 0.0001

Data Index: 1301131 (The Pile)

Max Activating Token Index: 44

Text #13

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.2812. Min Act: -0.0000

Data Index: 1825931 (The Pile)

Max Activating Token Index: 332

Text #14

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.3534. Min Act: -0.0000

Data Index: 1263516 (The Pile)

Max Activating Token Index: 575

Text #15

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.3477. Min Act: -0.0000

Data Index: 565105 (The Pile)

Max Activating Token Index: 773

Text #16

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.2415. Min Act: 0.0000

Data Index: 1213303 (The Pile)

Max Activating Token Index: 382

Text #17

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.5475. Min Act: -0.0000

Data Index: 635557 (The Pile)

Max Activating Token Index: 950

Text #18

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.4706. Min Act: -0.0000

Data Index: 596767 (The Pile)

Max Activating Token Index: 686

Text #19

Max Range: 1.7088. Min Range: -1.7088

Max Act: 1.2978. Min Act: -0.0000

Data Index: 106863 (The Pile)

Max Activating Token Index: 401