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

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

Neuron 1329 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1522. Min Act: -0.0001

Data Index: 1649760 (Python Code)

Max Activating Token Index: 313

Text #1

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1449. Min Act: -0.0001

Data Index: 212416 (C4 (Web Text))

Max Activating Token Index: 609

Text #2

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1343. Min Act: -0.0001

Data Index: 127498 (C4 (Web Text))

Max Activating Token Index: 5

Text #3

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1315. Min Act: -0.0001

Data Index: 445216 (C4 (Web Text))

Max Activating Token Index: 12

Text #4

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1302. Min Act: -0.0001

Data Index: 511144 (C4 (Web Text))

Max Activating Token Index: 725

Text #5

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1284. Min Act: -0.0001

Data Index: 1587271 (Python Code)

Max Activating Token Index: 3

Text #6

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1226. Min Act: -0.0001

Data Index: 710786 (C4 (Web Text))

Max Activating Token Index: 4

Text #7

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1162. Min Act: -0.0001

Data Index: 1050727 (C4 (Web Text))

Max Activating Token Index: 361

Text #8

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1168. Min Act: -0.0001

Data Index: 761596 (C4 (Web Text))

Max Activating Token Index: 36

Text #9

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1096. Min Act: -0.0001

Data Index: 1237346 (C4 (Web Text))

Max Activating Token Index: 514

Text #10

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1181. Min Act: -0.0001

Data Index: 1100926 (C4 (Web Text))

Max Activating Token Index: 914

Text #11

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1113. Min Act: -0.0001

Data Index: 850852 (C4 (Web Text))

Max Activating Token Index: 451

Text #12

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1124. Min Act: -0.0001

Data Index: 1130442 (C4 (Web Text))

Max Activating Token Index: 996

Text #13

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1137. Min Act: -0.0001

Data Index: 276946 (C4 (Web Text))

Max Activating Token Index: 592

Text #14

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1116. Min Act: -0.0001

Data Index: 439600 (C4 (Web Text))

Max Activating Token Index: 338

Text #15

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1117. Min Act: -0.0001

Data Index: 915721 (C4 (Web Text))

Max Activating Token Index: 398

Text #16

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1092. Min Act: -0.0001

Data Index: 974993 (C4 (Web Text))

Max Activating Token Index: 331

Text #17

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1130. Min Act: -0.0001

Data Index: 130648 (C4 (Web Text))

Max Activating Token Index: 587

Text #18

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1088. Min Act: -0.0001

Data Index: 1354394 (C4 (Web Text))

Max Activating Token Index: 944

Text #19

Max Range: 0.1522. Min Range: -0.1522

Max Act: 0.1062. Min Act: -0.0001

Data Index: 612399 (C4 (Web Text))

Max Activating Token Index: 721