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

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

Neuron 4793 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: 0.1570. Min Range: -0.1570

Max Act: 0.1570. Min Act: -0.0000

Data Index: 787287 (C4 (Web Text))

Max Activating Token Index: 230

Text #1

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1367. Min Act: -0.0000

Data Index: 156084 (C4 (Web Text))

Max Activating Token Index: 85

Text #2

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1544. Min Act: -0.0000

Data Index: 1299348 (C4 (Web Text))

Max Activating Token Index: 632

Text #3

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1197. Min Act: -0.0000

Data Index: 161239 (C4 (Web Text))

Max Activating Token Index: 168

Text #4

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1212. Min Act: -0.0000

Data Index: 114433 (C4 (Web Text))

Max Activating Token Index: 498

Text #5

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1219. Min Act: -0.0000

Data Index: 883597 (C4 (Web Text))

Max Activating Token Index: 301

Text #6

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1258. Min Act: -0.0001

Data Index: 102170 (C4 (Web Text))

Max Activating Token Index: 293

Text #7

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1169. Min Act: -0.0000

Data Index: 1248633 (C4 (Web Text))

Max Activating Token Index: 554

Text #8

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1123. Min Act: -0.0000

Data Index: 830357 (C4 (Web Text))

Max Activating Token Index: 501

Text #9

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1202. Min Act: -0.0001

Data Index: 29488 (C4 (Web Text))

Max Activating Token Index: 496

Text #10

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1190. Min Act: -0.0000

Data Index: 1178312 (C4 (Web Text))

Max Activating Token Index: 108

Text #11

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1195. Min Act: -0.0001

Data Index: 483801 (C4 (Web Text))

Max Activating Token Index: 417

Text #12

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1187. Min Act: -0.0000

Data Index: 1075641 (C4 (Web Text))

Max Activating Token Index: 597

Text #13

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1186. Min Act: -0.0000

Data Index: 1082417 (C4 (Web Text))

Max Activating Token Index: 579

Text #14

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1143. Min Act: -0.0000

Data Index: 546915 (C4 (Web Text))

Max Activating Token Index: 619

Text #15

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1145. Min Act: -0.0001

Data Index: 849734 (C4 (Web Text))

Max Activating Token Index: 436

Text #16

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1052. Min Act: -0.0000

Data Index: 1547848 (Python Code)

Max Activating Token Index: 762

Text #17

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1217. Min Act: -0.0000

Data Index: 629260 (C4 (Web Text))

Max Activating Token Index: 844

Text #18

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1135. Min Act: -0.0000

Data Index: 697927 (C4 (Web Text))

Max Activating Token Index: 888

Text #19

Max Range: 0.1570. Min Range: -0.1570

Max Act: 0.1126. Min Act: -0.0000

Data Index: 822668 (C4 (Web Text))

Max Activating Token Index: 585