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

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

Neuron 2854 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.6327. Min Act: -0.0000

Data Index: 1292884 (C4 (Web Text))

Max Activating Token Index: 859

Text #1

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.4892. Min Act: -0.0000

Data Index: 15655 (C4 (Web Text))

Max Activating Token Index: 105

Text #2

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.4163. Min Act: -0.0000

Data Index: 1560594 (Python Code)

Max Activating Token Index: 404

Text #3

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.4091. Min Act: -0.0000

Data Index: 1225696 (C4 (Web Text))

Max Activating Token Index: 636

Text #4

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.4166. Min Act: -0.0000

Data Index: 1170987 (C4 (Web Text))

Max Activating Token Index: 906

Text #5

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.4121. Min Act: -0.0000

Data Index: 1265050 (C4 (Web Text))

Max Activating Token Index: 1014

Text #6

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3929. Min Act: -0.0000

Data Index: 1334312 (C4 (Web Text))

Max Activating Token Index: 932

Text #7

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3784. Min Act: -0.0000

Data Index: 900495 (C4 (Web Text))

Max Activating Token Index: 335

Text #8

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3520. Min Act: -0.0000

Data Index: 1175722 (C4 (Web Text))

Max Activating Token Index: 654

Text #9

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3771. Min Act: -0.0000

Data Index: 466911 (C4 (Web Text))

Max Activating Token Index: 318

Text #10

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3772. Min Act: -0.0000

Data Index: 1614108 (Python Code)

Max Activating Token Index: 310

Text #11

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3398. Min Act: -0.0000

Data Index: 434402 (C4 (Web Text))

Max Activating Token Index: 981

Text #12

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3490. Min Act: -0.0000

Data Index: 94864 (C4 (Web Text))

Max Activating Token Index: 560

Text #13

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3508. Min Act: -0.0000

Data Index: 452498 (C4 (Web Text))

Max Activating Token Index: 309

Text #14

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3616. Min Act: -0.0000

Data Index: 832546 (C4 (Web Text))

Max Activating Token Index: 940

Text #15

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3625. Min Act: -0.0000

Data Index: 500175 (C4 (Web Text))

Max Activating Token Index: 613

Text #16

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3470. Min Act: -0.0000

Data Index: 860398 (C4 (Web Text))

Max Activating Token Index: 45

Text #17

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3202. Min Act: -0.0000

Data Index: 904201 (C4 (Web Text))

Max Activating Token Index: 871

Text #18

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3934. Min Act: -0.0000

Data Index: 1447825 (Python Code)

Max Activating Token Index: 213

Text #19

Max Range: 0.6327. Min Range: -0.6327

Max Act: 0.3262. Min Act: -0.0000

Data Index: 750064 (C4 (Web Text))

Max Activating Token Index: 968