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

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

Neuron 673 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: 0.1444. Min Range: -0.1444

Max Act: 0.1444. Min Act: -0.0001

Data Index: 609418 (C4 (Web Text))

Max Activating Token Index: 60

Text #1

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1358. Min Act: -0.0000

Data Index: 8781 (C4 (Web Text))

Max Activating Token Index: 791

Text #2

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1337. Min Act: -0.0000

Data Index: 144278 (C4 (Web Text))

Max Activating Token Index: 830

Text #3

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1290. Min Act: -0.0000

Data Index: 328670 (C4 (Web Text))

Max Activating Token Index: 380

Text #4

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1312. Min Act: -0.0000

Data Index: 969565 (C4 (Web Text))

Max Activating Token Index: 61

Text #5

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1239. Min Act: -0.0000

Data Index: 1215796 (C4 (Web Text))

Max Activating Token Index: 56

Text #6

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1228. Min Act: -0.0000

Data Index: 296508 (C4 (Web Text))

Max Activating Token Index: 934

Text #7

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1223. Min Act: -0.0000

Data Index: 667217 (C4 (Web Text))

Max Activating Token Index: 892

Text #8

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1178. Min Act: -0.0000

Data Index: 827473 (C4 (Web Text))

Max Activating Token Index: 899

Text #9

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1198. Min Act: -0.0000

Data Index: 683488 (C4 (Web Text))

Max Activating Token Index: 159

Text #10

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1226. Min Act: -0.0000

Data Index: 1562320 (Python Code)

Max Activating Token Index: 218

Text #11

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1189. Min Act: -0.0000

Data Index: 1024116 (C4 (Web Text))

Max Activating Token Index: 448

Text #12

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1186. Min Act: -0.0000

Data Index: 498660 (C4 (Web Text))

Max Activating Token Index: 773

Text #13

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1180. Min Act: -0.0000

Data Index: 1275366 (C4 (Web Text))

Max Activating Token Index: 74

Text #14

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1090. Min Act: -0.0000

Data Index: 28101 (C4 (Web Text))

Max Activating Token Index: 279

Text #15

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1171. Min Act: -0.0000

Data Index: 676310 (C4 (Web Text))

Max Activating Token Index: 445

Text #16

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1157. Min Act: -0.0000

Data Index: 879259 (C4 (Web Text))

Max Activating Token Index: 587

Text #17

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1200. Min Act: -0.0000

Data Index: 1180213 (C4 (Web Text))

Max Activating Token Index: 810

Text #18

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1111. Min Act: -0.0000

Data Index: 1395436 (Python Code)

Max Activating Token Index: 365

Text #19

Max Range: 0.1444. Min Range: -0.1444

Max Act: 0.1237. Min Act: -0.0000

Data Index: 1339077 (C4 (Web Text))

Max Activating Token Index: 214