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

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

Neuron 4793 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1339. Min Act: -0.0001

Data Index: 128314 (C4 (Web Text))

Max Activating Token Index: 246

Text #1

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1267. Min Act: -0.0001

Data Index: 206755 (C4 (Web Text))

Max Activating Token Index: 238

Text #2

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1250. Min Act: -0.0001

Data Index: 474176 (C4 (Web Text))

Max Activating Token Index: 282

Text #3

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1222. Min Act: -0.0001

Data Index: 376540 (C4 (Web Text))

Max Activating Token Index: 590

Text #4

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1224. Min Act: -0.0001

Data Index: 614212 (C4 (Web Text))

Max Activating Token Index: 539

Text #5

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1195. Min Act: -0.0001

Data Index: 946342 (C4 (Web Text))

Max Activating Token Index: 462

Text #6

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1178. Min Act: -0.0001

Data Index: 440816 (C4 (Web Text))

Max Activating Token Index: 588

Text #7

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1120. Min Act: -0.0001

Data Index: 517880 (C4 (Web Text))

Max Activating Token Index: 509

Text #8

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1121. Min Act: -0.0001

Data Index: 239610 (C4 (Web Text))

Max Activating Token Index: 604

Text #9

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1101. Min Act: -0.0001

Data Index: 673461 (C4 (Web Text))

Max Activating Token Index: 838

Text #10

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1160. Min Act: -0.0001

Data Index: 1130411 (C4 (Web Text))

Max Activating Token Index: 649

Text #11

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1052. Min Act: -0.0001

Data Index: 981637 (C4 (Web Text))

Max Activating Token Index: 756

Text #12

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1063. Min Act: -0.0001

Data Index: 289202 (C4 (Web Text))

Max Activating Token Index: 463

Text #13

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.0991. Min Act: -0.0001

Data Index: 374362 (C4 (Web Text))

Max Activating Token Index: 696

Text #14

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.0985. Min Act: -0.0001

Data Index: 959526 (C4 (Web Text))

Max Activating Token Index: 370

Text #15

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.0993. Min Act: -0.0001

Data Index: 1007049 (C4 (Web Text))

Max Activating Token Index: 750

Text #16

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.1032. Min Act: -0.0001

Data Index: 340122 (C4 (Web Text))

Max Activating Token Index: 546

Text #17

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.0949. Min Act: -0.0001

Data Index: 449966 (C4 (Web Text))

Max Activating Token Index: 580

Text #18

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.0936. Min Act: -0.0001

Data Index: 583235 (C4 (Web Text))

Max Activating Token Index: 771

Text #19

Max Range: 0.1339. Min Range: -0.1339

Max Act: 0.0940. Min Act: -0.0001

Data Index: 318403 (C4 (Web Text))

Max Activating Token Index: 83