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

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

Neuron 38 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0245. Min Range: -1.0245

Max Act: 1.0245. Min Act: -0.0000

Data Index: 294040 (C4 (Web Text))

Max Activating Token Index: 558

Text #1

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.9866. Min Act: -0.0000

Data Index: 1649675 (Python Code)

Max Activating Token Index: 369

Text #2

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.9276. Min Act: -0.0000

Data Index: 1133462 (C4 (Web Text))

Max Activating Token Index: 725

Text #3

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.8194. Min Act: -0.0000

Data Index: 1413989 (Python Code)

Max Activating Token Index: 246

Text #4

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7952. Min Act: -0.0000

Data Index: 1432730 (Python Code)

Max Activating Token Index: 195

Text #5

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.8311. Min Act: -0.0000

Data Index: 893438 (C4 (Web Text))

Max Activating Token Index: 383

Text #6

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.8523. Min Act: -0.0000

Data Index: 1381726 (Python Code)

Max Activating Token Index: 399

Text #7

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.8134. Min Act: -0.0000

Data Index: 830508 (C4 (Web Text))

Max Activating Token Index: 844

Text #8

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7810. Min Act: -0.0000

Data Index: 1017562 (C4 (Web Text))

Max Activating Token Index: 219

Text #9

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7577. Min Act: -0.0000

Data Index: 249806 (C4 (Web Text))

Max Activating Token Index: 282

Text #10

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7696. Min Act: -0.0000

Data Index: 478771 (C4 (Web Text))

Max Activating Token Index: 381

Text #11

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7800. Min Act: -0.0000

Data Index: 643204 (C4 (Web Text))

Max Activating Token Index: 883

Text #12

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7302. Min Act: -0.0000

Data Index: 609570 (C4 (Web Text))

Max Activating Token Index: 44

Text #13

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7975. Min Act: -0.0000

Data Index: 1543391 (Python Code)

Max Activating Token Index: 974

Text #14

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7705. Min Act: -0.0000

Data Index: 885334 (C4 (Web Text))

Max Activating Token Index: 855

Text #15

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7188. Min Act: -0.0000

Data Index: 527129 (C4 (Web Text))

Max Activating Token Index: 698

Text #16

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7386. Min Act: -0.0000

Data Index: 944727 (C4 (Web Text))

Max Activating Token Index: 742

Text #17

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7868. Min Act: -0.0000

Data Index: 932729 (C4 (Web Text))

Max Activating Token Index: 123

Text #18

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7319. Min Act: -0.0000

Data Index: 1645729 (Python Code)

Max Activating Token Index: 953

Text #19

Max Range: 1.0245. Min Range: -1.0245

Max Act: 0.7029. Min Act: -0.0000

Data Index: 1167721 (C4 (Web Text))

Max Activating Token Index: 203