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

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

Neuron 4822 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.7859. Min Act: -0.0000

Data Index: 372180 (C4 (Web Text))

Max Activating Token Index: 208

Text #1

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.7674. Min Act: -0.0000

Data Index: 1164819 (C4 (Web Text))

Max Activating Token Index: 218

Text #2

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.6169. Min Act: -0.0000

Data Index: 348524 (C4 (Web Text))

Max Activating Token Index: 799

Text #3

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.5800. Min Act: -0.0000

Data Index: 1178763 (C4 (Web Text))

Max Activating Token Index: 855

Text #4

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.5274. Min Act: -0.0000

Data Index: 261386 (C4 (Web Text))

Max Activating Token Index: 351

Text #5

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.5254. Min Act: -0.0000

Data Index: 746829 (C4 (Web Text))

Max Activating Token Index: 376

Text #6

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.5160. Min Act: -0.0000

Data Index: 1144031 (C4 (Web Text))

Max Activating Token Index: 535

Text #7

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.5102. Min Act: -0.0000

Data Index: 675956 (C4 (Web Text))

Max Activating Token Index: 79

Text #8

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4846. Min Act: -0.0000

Data Index: 527939 (C4 (Web Text))

Max Activating Token Index: 52

Text #9

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4479. Min Act: -0.0000

Data Index: 903589 (C4 (Web Text))

Max Activating Token Index: 284

Text #10

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4491. Min Act: -0.0000

Data Index: 1185489 (C4 (Web Text))

Max Activating Token Index: 266

Text #11

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4408. Min Act: -0.0000

Data Index: 1620007 (Python Code)

Max Activating Token Index: 109

Text #12

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4320. Min Act: -0.0000

Data Index: 196476 (C4 (Web Text))

Max Activating Token Index: 709

Text #13

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4674. Min Act: -0.0000

Data Index: 1599599 (Python Code)

Max Activating Token Index: 979

Text #14

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4382. Min Act: -0.0000

Data Index: 970707 (C4 (Web Text))

Max Activating Token Index: 493

Text #15

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4179. Min Act: -0.0000

Data Index: 805711 (C4 (Web Text))

Max Activating Token Index: 156

Text #16

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4235. Min Act: -0.0000

Data Index: 1054174 (C4 (Web Text))

Max Activating Token Index: 470

Text #17

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4166. Min Act: -0.0000

Data Index: 1523140 (Python Code)

Max Activating Token Index: 242

Text #18

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.4155. Min Act: -0.0000

Data Index: 1400697 (Python Code)

Max Activating Token Index: 741

Text #19

Max Range: 0.7859. Min Range: -0.7859

Max Act: 0.3997. Min Act: -0.0000

Data Index: 342140 (C4 (Web Text))

Max Activating Token Index: 312