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

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

Neuron 415 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0461. Min Act: -0.0006

Data Index: 496045 (C4 (Web Text))

Max Activating Token Index: 575

Text #1

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0453. Min Act: -0.0006

Data Index: 929148 (C4 (Web Text))

Max Activating Token Index: 459

Text #2

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0488. Min Act: -0.0006

Data Index: 622763 (C4 (Web Text))

Max Activating Token Index: 413

Text #3

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0437. Min Act: -0.0006

Data Index: 1099307 (C4 (Web Text))

Max Activating Token Index: 959

Text #4

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0419. Min Act: -0.0006

Data Index: 1413283 (Python Code)

Max Activating Token Index: 802

Text #5

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0417. Min Act: -0.0006

Data Index: 130642 (C4 (Web Text))

Max Activating Token Index: 894

Text #6

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0448. Min Act: -0.0006

Data Index: 530280 (C4 (Web Text))

Max Activating Token Index: 104

Text #7

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0396. Min Act: -0.0006

Data Index: 749389 (C4 (Web Text))

Max Activating Token Index: 701

Text #8

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0398. Min Act: -0.0006

Data Index: 797456 (C4 (Web Text))

Max Activating Token Index: 653

Text #9

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0361. Min Act: -0.0006

Data Index: 705953 (C4 (Web Text))

Max Activating Token Index: 661

Text #10

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0409. Min Act: -0.0006

Data Index: 262373 (C4 (Web Text))

Max Activating Token Index: 557

Text #11

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0369. Min Act: -0.0006

Data Index: 1579326 (Python Code)

Max Activating Token Index: 532

Text #12

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0385. Min Act: -0.0006

Data Index: 1303400 (C4 (Web Text))

Max Activating Token Index: 462

Text #13

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0377. Min Act: -0.0006

Data Index: 748821 (C4 (Web Text))

Max Activating Token Index: 596

Text #14

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0360. Min Act: -0.0006

Data Index: 1466256 (Python Code)

Max Activating Token Index: 404

Text #15

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0389. Min Act: -0.0006

Data Index: 260489 (C4 (Web Text))

Max Activating Token Index: 776

Text #16

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0389. Min Act: -0.0006

Data Index: 279383 (C4 (Web Text))

Max Activating Token Index: 797

Text #17

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0387. Min Act: -0.0006

Data Index: 673780 (C4 (Web Text))

Max Activating Token Index: 892

Text #18

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0369. Min Act: -0.0006

Data Index: 1356526 (C4 (Web Text))

Max Activating Token Index: 763

Text #19

Max Range: 0.0488. Min Range: -0.0488

Max Act: 0.0384. Min Act: -0.0006

Data Index: 938228 (C4 (Web Text))

Max Activating Token Index: 1021