Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1959 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.9379. Min Act: -0.0000

Data Index: 911686 (C4 (Web Text))

Max Activating Token Index: 136

Text #1

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.7244. Min Act: -0.0000

Data Index: 209728 (C4 (Web Text))

Max Activating Token Index: 771

Text #2

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.7437. Min Act: -0.0000

Data Index: 1370793 (Python Code)

Max Activating Token Index: 125

Text #3

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.7118. Min Act: -0.0000

Data Index: 1322840 (C4 (Web Text))

Max Activating Token Index: 88

Text #4

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6420. Min Act: -0.0000

Data Index: 216952 (C4 (Web Text))

Max Activating Token Index: 857

Text #5

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6654. Min Act: -0.0000

Data Index: 1325861 (C4 (Web Text))

Max Activating Token Index: 91

Text #6

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6410. Min Act: -0.0000

Data Index: 1076202 (C4 (Web Text))

Max Activating Token Index: 951

Text #7

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6510. Min Act: -0.0000

Data Index: 1247631 (C4 (Web Text))

Max Activating Token Index: 675

Text #8

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6391. Min Act: -0.0000

Data Index: 113248 (C4 (Web Text))

Max Activating Token Index: 960

Text #9

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6351. Min Act: -0.0000

Data Index: 638046 (C4 (Web Text))

Max Activating Token Index: 527

Text #10

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6506. Min Act: -0.0000

Data Index: 11722 (C4 (Web Text))

Max Activating Token Index: 642

Text #11

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6309. Min Act: -0.0000

Data Index: 841034 (C4 (Web Text))

Max Activating Token Index: 890

Text #12

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6375. Min Act: -0.0000

Data Index: 422956 (C4 (Web Text))

Max Activating Token Index: 832

Text #13

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6511. Min Act: -0.0000

Data Index: 1480413 (Python Code)

Max Activating Token Index: 541

Text #14

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6076. Min Act: -0.0000

Data Index: 279537 (C4 (Web Text))

Max Activating Token Index: 201

Text #15

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6201. Min Act: -0.0000

Data Index: 574168 (C4 (Web Text))

Max Activating Token Index: 396

Text #16

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6834. Min Act: -0.0000

Data Index: 1388276 (Python Code)

Max Activating Token Index: 230

Text #17

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6365. Min Act: -0.0000

Data Index: 595354 (C4 (Web Text))

Max Activating Token Index: 744

Text #18

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.5713. Min Act: -0.0000

Data Index: 878487 (C4 (Web Text))

Max Activating Token Index: 891

Text #19

Max Range: 0.9379. Min Range: -0.9379

Max Act: 0.6112. Min Act: -0.0000

Data Index: 730913 (C4 (Web Text))

Max Activating Token Index: 265