Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: The Pile

Neuron 29 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 3.3470. Min Range: -3.3470

Max Act: 3.2895. Min Act: -0.0000

Data Index: 770085 (The Pile)

Max Activating Token Index: 824

Text #1

Max Range: 3.3470. Min Range: -3.3470

Max Act: 3.2377. Min Act: -0.0000

Data Index: 1176331 (The Pile)

Max Activating Token Index: 689

Text #2

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.8217. Min Act: -0.0000

Data Index: 1903357 (The Pile)

Max Activating Token Index: 668

Text #3

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.9379. Min Act: -0.0001

Data Index: 566462 (The Pile)

Max Activating Token Index: 73

Text #4

Max Range: 3.3470. Min Range: -3.3470

Max Act: 3.3470. Min Act: -0.0000

Data Index: 364151 (The Pile)

Max Activating Token Index: 398

Text #5

Max Range: 3.3470. Min Range: -3.3470

Max Act: 3.0418. Min Act: -0.0000

Data Index: 396458 (The Pile)

Max Activating Token Index: 691

Text #6

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.9385. Min Act: -0.0000

Data Index: 1981606 (The Pile)

Max Activating Token Index: 1014

Text #7

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.9485. Min Act: -0.0000

Data Index: 1267275 (The Pile)

Max Activating Token Index: 574

Text #8

Max Range: 3.3470. Min Range: -3.3470

Max Act: 3.0167. Min Act: -0.0000

Data Index: 1632849 (The Pile)

Max Activating Token Index: 654

Text #9

Max Range: 3.3470. Min Range: -3.3470

Max Act: 3.0303. Min Act: -0.0000

Data Index: 940077 (The Pile)

Max Activating Token Index: 265

Text #10

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.8222. Min Act: -0.0000

Data Index: 1303290 (The Pile)

Max Activating Token Index: 624

Text #11

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.8748. Min Act: -0.0000

Data Index: 159850 (The Pile)

Max Activating Token Index: 815

Text #12

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.5534. Min Act: -0.0000

Data Index: 1019537 (The Pile)

Max Activating Token Index: 782

Text #13

Max Range: 3.3470. Min Range: -3.3470

Max Act: 3.0749. Min Act: -0.0000

Data Index: 699871 (The Pile)

Max Activating Token Index: 658

Text #14

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.9752. Min Act: -0.0000

Data Index: 334351 (The Pile)

Max Activating Token Index: 878

Text #15

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.9177. Min Act: -0.0000

Data Index: 339793 (The Pile)

Max Activating Token Index: 924

Text #16

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.4773. Min Act: -0.0000

Data Index: 271559 (The Pile)

Max Activating Token Index: 951

Text #17

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.8553. Min Act: -0.0000

Data Index: 1707847 (The Pile)

Max Activating Token Index: 579

Text #18

Max Range: 3.3470. Min Range: -3.3470

Max Act: 3.1392. Min Act: -0.0000

Data Index: 1973238 (The Pile)

Max Activating Token Index: 697

Text #19

Max Range: 3.3470. Min Range: -3.3470

Max Act: 2.8351. Min Act: -0.0000

Data Index: 871231 (The Pile)

Max Activating Token Index: 884