Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1493 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1358. Min Act: -0.0005

Data Index: 700141 (C4 (Web Text))

Max Activating Token Index: 582

Text #1

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1305. Min Act: -0.0005

Data Index: 802729 (C4 (Web Text))

Max Activating Token Index: 703

Text #2

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1319. Min Act: -0.0005

Data Index: 135283 (C4 (Web Text))

Max Activating Token Index: 733

Text #3

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1321. Min Act: -0.0005

Data Index: 1493474 (Python Code)

Max Activating Token Index: 666

Text #4

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1300. Min Act: -0.0005

Data Index: 1055948 (C4 (Web Text))

Max Activating Token Index: 520

Text #5

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1245. Min Act: -0.0005

Data Index: 1083901 (C4 (Web Text))

Max Activating Token Index: 457

Text #6

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1222. Min Act: -0.0005

Data Index: 1499124 (Python Code)

Max Activating Token Index: 827

Text #7

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1264. Min Act: -0.0005

Data Index: 1421003 (Python Code)

Max Activating Token Index: 528

Text #8

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1214. Min Act: -0.0005

Data Index: 1476285 (Python Code)

Max Activating Token Index: 572

Text #9

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1261. Min Act: -0.0005

Data Index: 1065708 (C4 (Web Text))

Max Activating Token Index: 866

Text #10

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1285. Min Act: -0.0005

Data Index: 640141 (C4 (Web Text))

Max Activating Token Index: 588

Text #11

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1188. Min Act: -0.0005

Data Index: 1521438 (Python Code)

Max Activating Token Index: 668

Text #12

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1227. Min Act: -0.0005

Data Index: 863367 (C4 (Web Text))

Max Activating Token Index: 747

Text #13

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1234. Min Act: -0.0005

Data Index: 1389791 (Python Code)

Max Activating Token Index: 919

Text #14

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1296. Min Act: -0.0005

Data Index: 1130375 (C4 (Web Text))

Max Activating Token Index: 774

Text #15

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1273. Min Act: -0.0005

Data Index: 368057 (C4 (Web Text))

Max Activating Token Index: 835

Text #16

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1278. Min Act: -0.0005

Data Index: 1570071 (Python Code)

Max Activating Token Index: 634

Text #17

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1152. Min Act: -0.0005

Data Index: 1586063 (Python Code)

Max Activating Token Index: 575

Text #18

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1223. Min Act: -0.0005

Data Index: 1495684 (Python Code)

Max Activating Token Index: 655

Text #19

Max Range: 0.1358. Min Range: -0.1358

Max Act: 0.1265. Min Act: -0.0005

Data Index: 761528 (C4 (Web Text))

Max Activating Token Index: 730