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

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

Neuron 479 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1701. Min Act: -0.0002

Data Index: 1367561 (Python Code)

Max Activating Token Index: 664

Text #1

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1495. Min Act: -0.0002

Data Index: 777964 (C4 (Web Text))

Max Activating Token Index: 622

Text #2

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1496. Min Act: -0.0002

Data Index: 365722 (C4 (Web Text))

Max Activating Token Index: 493

Text #3

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1338. Min Act: -0.0002

Data Index: 1167988 (C4 (Web Text))

Max Activating Token Index: 808

Text #4

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1452. Min Act: -0.0002

Data Index: 1053440 (C4 (Web Text))

Max Activating Token Index: 643

Text #5

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1266. Min Act: -0.0002

Data Index: 1358199 (C4 (Web Text))

Max Activating Token Index: 267

Text #6

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1348. Min Act: -0.0002

Data Index: 326617 (C4 (Web Text))

Max Activating Token Index: 590

Text #7

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1356. Min Act: -0.0002

Data Index: 1114521 (C4 (Web Text))

Max Activating Token Index: 810

Text #8

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1244. Min Act: -0.0002

Data Index: 1482404 (Python Code)

Max Activating Token Index: 423

Text #9

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1310. Min Act: -0.0002

Data Index: 1630134 (Python Code)

Max Activating Token Index: 366

Text #10

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1303. Min Act: -0.0002

Data Index: 684498 (C4 (Web Text))

Max Activating Token Index: 339

Text #11

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1276. Min Act: -0.0002

Data Index: 1547097 (Python Code)

Max Activating Token Index: 629

Text #12

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1363. Min Act: -0.0002

Data Index: 1656527 (Python Code)

Max Activating Token Index: 211

Text #13

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1297. Min Act: -0.0002

Data Index: 952633 (C4 (Web Text))

Max Activating Token Index: 24

Text #14

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1378. Min Act: -0.0002

Data Index: 483388 (C4 (Web Text))

Max Activating Token Index: 109

Text #15

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1365. Min Act: -0.0002

Data Index: 650557 (C4 (Web Text))

Max Activating Token Index: 545

Text #16

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1241. Min Act: -0.0002

Data Index: 1155885 (C4 (Web Text))

Max Activating Token Index: 850

Text #17

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1319. Min Act: -0.0002

Data Index: 899757 (C4 (Web Text))

Max Activating Token Index: 395

Text #18

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1215. Min Act: -0.0002

Data Index: 898597 (C4 (Web Text))

Max Activating Token Index: 488

Text #19

Max Range: 0.1701. Min Range: -0.1701

Max Act: 0.1259. Min Act: -0.0002

Data Index: 741421 (C4 (Web Text))

Max Activating Token Index: 601