Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3475 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1497. Min Act: -0.0001

Data Index: 539556 (C4 (Web Text))

Max Activating Token Index: 215

Text #1

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1511. Min Act: -0.0001

Data Index: 197574 (C4 (Web Text))

Max Activating Token Index: 978

Text #2

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1409. Min Act: -0.0001

Data Index: 448090 (C4 (Web Text))

Max Activating Token Index: 832

Text #3

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1338. Min Act: -0.0001

Data Index: 1235235 (C4 (Web Text))

Max Activating Token Index: 927

Text #4

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1212. Min Act: -0.0001

Data Index: 1241389 (C4 (Web Text))

Max Activating Token Index: 928

Text #5

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1381. Min Act: -0.0001

Data Index: 357347 (C4 (Web Text))

Max Activating Token Index: 938

Text #6

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1368. Min Act: -0.0001

Data Index: 1202839 (C4 (Web Text))

Max Activating Token Index: 337

Text #7

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1110. Min Act: -0.0001

Data Index: 18469 (C4 (Web Text))

Max Activating Token Index: 612

Text #8

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1335. Min Act: -0.0001

Data Index: 942490 (C4 (Web Text))

Max Activating Token Index: 134

Text #9

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1160. Min Act: -0.0001

Data Index: 1146279 (C4 (Web Text))

Max Activating Token Index: 710

Text #10

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1263. Min Act: -0.0001

Data Index: 257513 (C4 (Web Text))

Max Activating Token Index: 891

Text #11

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1206. Min Act: -0.0001

Data Index: 574417 (C4 (Web Text))

Max Activating Token Index: 116

Text #12

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1140. Min Act: -0.0001

Data Index: 1024606 (C4 (Web Text))

Max Activating Token Index: 654

Text #13

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1085. Min Act: -0.0001

Data Index: 1351847 (C4 (Web Text))

Max Activating Token Index: 720

Text #14

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1118. Min Act: -0.0001

Data Index: 1458341 (Python Code)

Max Activating Token Index: 566

Text #15

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1096. Min Act: -0.0001

Data Index: 756123 (C4 (Web Text))

Max Activating Token Index: 355

Text #16

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1164. Min Act: -0.0001

Data Index: 1077149 (C4 (Web Text))

Max Activating Token Index: 131

Text #17

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1085. Min Act: -0.0001

Data Index: 884124 (C4 (Web Text))

Max Activating Token Index: 364

Text #18

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1078. Min Act: -0.0001

Data Index: 1577224 (Python Code)

Max Activating Token Index: 869

Text #19

Max Range: 0.1511. Min Range: -0.1511

Max Act: 0.1070. Min Act: -0.0001

Data Index: 1190635 (C4 (Web Text))

Max Activating Token Index: 615