Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1888 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0457. Min Act: -0.0003

Data Index: 883251 (C4 (Web Text))

Max Activating Token Index: 734

Text #1

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0412. Min Act: -0.0003

Data Index: 1510793 (Python Code)

Max Activating Token Index: 520

Text #2

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0410. Min Act: -0.0002

Data Index: 770850 (C4 (Web Text))

Max Activating Token Index: 495

Text #3

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0405. Min Act: -0.0002

Data Index: 554783 (C4 (Web Text))

Max Activating Token Index: 1011

Text #4

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0393. Min Act: -0.0002

Data Index: 318773 (C4 (Web Text))

Max Activating Token Index: 874

Text #5

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0379. Min Act: -0.0003

Data Index: 647636 (C4 (Web Text))

Max Activating Token Index: 536

Text #6

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0389. Min Act: -0.0002

Data Index: 345946 (C4 (Web Text))

Max Activating Token Index: 860

Text #7

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0371. Min Act: -0.0003

Data Index: 494610 (C4 (Web Text))

Max Activating Token Index: 965

Text #8

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0373. Min Act: -0.0003

Data Index: 1519503 (Python Code)

Max Activating Token Index: 419

Text #9

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0377. Min Act: -0.0003

Data Index: 679504 (C4 (Web Text))

Max Activating Token Index: 651

Text #10

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0358. Min Act: -0.0003

Data Index: 3083 (C4 (Web Text))

Max Activating Token Index: 314

Text #11

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0360. Min Act: -0.0003

Data Index: 86550 (C4 (Web Text))

Max Activating Token Index: 736

Text #12

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0350. Min Act: -0.0003

Data Index: 1417365 (Python Code)

Max Activating Token Index: 837

Text #13

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0351. Min Act: -0.0003

Data Index: 315264 (C4 (Web Text))

Max Activating Token Index: 299

Text #14

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0361. Min Act: -0.0004

Data Index: 89788 (C4 (Web Text))

Max Activating Token Index: 769

Text #15

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0358. Min Act: -0.0003

Data Index: 61019 (C4 (Web Text))

Max Activating Token Index: 907

Text #16

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0350. Min Act: -0.0003

Data Index: 136472 (C4 (Web Text))

Max Activating Token Index: 167

Text #17

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0343. Min Act: -0.0003

Data Index: 222721 (C4 (Web Text))

Max Activating Token Index: 797

Text #18

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0348. Min Act: -0.0003

Data Index: 501956 (C4 (Web Text))

Max Activating Token Index: 343

Text #19

Max Range: 0.0457. Min Range: -0.0457

Max Act: 0.0344. Min Act: -0.0002

Data Index: 718045 (C4 (Web Text))

Max Activating Token Index: 717