Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3200 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.6061. Min Act: -0.0000

Data Index: 525330 (C4 (Web Text))

Max Activating Token Index: 457

Text #1

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.4571. Min Act: -0.0000

Data Index: 1274510 (C4 (Web Text))

Max Activating Token Index: 855

Text #2

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.4548. Min Act: -0.0000

Data Index: 606247 (C4 (Web Text))

Max Activating Token Index: 192

Text #3

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.4859. Min Act: -0.0000

Data Index: 1115480 (C4 (Web Text))

Max Activating Token Index: 574

Text #4

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.4435. Min Act: -0.0000

Data Index: 160606 (C4 (Web Text))

Max Activating Token Index: 659

Text #5

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.4321. Min Act: -0.0000

Data Index: 69436 (C4 (Web Text))

Max Activating Token Index: 935

Text #6

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.4365. Min Act: -0.0000

Data Index: 702828 (C4 (Web Text))

Max Activating Token Index: 387

Text #7

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3944. Min Act: -0.0000

Data Index: 281533 (C4 (Web Text))

Max Activating Token Index: 434

Text #8

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.4137. Min Act: -0.0000

Data Index: 1294355 (C4 (Web Text))

Max Activating Token Index: 78

Text #9

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3944. Min Act: -0.0000

Data Index: 1602680 (Python Code)

Max Activating Token Index: 567

Text #10

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3981. Min Act: -0.0000

Data Index: 653318 (C4 (Web Text))

Max Activating Token Index: 751

Text #11

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3823. Min Act: -0.0000

Data Index: 470260 (C4 (Web Text))

Max Activating Token Index: 531

Text #12

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3790. Min Act: -0.0000

Data Index: 619428 (C4 (Web Text))

Max Activating Token Index: 280

Text #13

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3755. Min Act: -0.0000

Data Index: 315152 (C4 (Web Text))

Max Activating Token Index: 266

Text #14

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3595. Min Act: -0.0000

Data Index: 1374020 (Python Code)

Max Activating Token Index: 622

Text #15

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3457. Min Act: -0.0000

Data Index: 728434 (C4 (Web Text))

Max Activating Token Index: 108

Text #16

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3782. Min Act: -0.0000

Data Index: 1014896 (C4 (Web Text))

Max Activating Token Index: 198

Text #17

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3555. Min Act: -0.0000

Data Index: 1390743 (Python Code)

Max Activating Token Index: 353

Text #18

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3446. Min Act: -0.0000

Data Index: 1041899 (C4 (Web Text))

Max Activating Token Index: 1011

Text #19

Max Range: 0.6061. Min Range: -0.6061

Max Act: 0.3401. Min Act: -0.0000

Data Index: 234012 (C4 (Web Text))

Max Activating Token Index: 766