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

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

Neuron 1822 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.7490. Min Act: -0.0000

Data Index: 162052 (C4 (Web Text))

Max Activating Token Index: 706

Text #1

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.6734. Min Act: -0.0000

Data Index: 143313 (C4 (Web Text))

Max Activating Token Index: 799

Text #2

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.6151. Min Act: -0.0000

Data Index: 80500 (C4 (Web Text))

Max Activating Token Index: 435

Text #3

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.6285. Min Act: -0.0000

Data Index: 1570322 (Python Code)

Max Activating Token Index: 559

Text #4

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5957. Min Act: -0.0000

Data Index: 1503711 (Python Code)

Max Activating Token Index: 716

Text #5

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.7200. Min Act: -0.0000

Data Index: 1151571 (C4 (Web Text))

Max Activating Token Index: 832

Text #6

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5481. Min Act: -0.0000

Data Index: 680565 (C4 (Web Text))

Max Activating Token Index: 494

Text #7

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5266. Min Act: -0.0000

Data Index: 1436063 (Python Code)

Max Activating Token Index: 666

Text #8

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5200. Min Act: -0.0000

Data Index: 82840 (C4 (Web Text))

Max Activating Token Index: 565

Text #9

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5357. Min Act: -0.0000

Data Index: 352637 (C4 (Web Text))

Max Activating Token Index: 906

Text #10

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5298. Min Act: -0.0000

Data Index: 535597 (C4 (Web Text))

Max Activating Token Index: 729

Text #11

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5259. Min Act: -0.0000

Data Index: 806977 (C4 (Web Text))

Max Activating Token Index: 698

Text #12

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5486. Min Act: -0.0000

Data Index: 695127 (C4 (Web Text))

Max Activating Token Index: 754

Text #13

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5290. Min Act: -0.0000

Data Index: 920538 (C4 (Web Text))

Max Activating Token Index: 852

Text #14

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.5203. Min Act: -0.0000

Data Index: 230939 (C4 (Web Text))

Max Activating Token Index: 383

Text #15

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.4815. Min Act: -0.0000

Data Index: 381980 (C4 (Web Text))

Max Activating Token Index: 69

Text #16

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.4876. Min Act: -0.0000

Data Index: 1529103 (Python Code)

Max Activating Token Index: 863

Text #17

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.4837. Min Act: -0.0000

Data Index: 194700 (C4 (Web Text))

Max Activating Token Index: 900

Text #18

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.4985. Min Act: -0.0000

Data Index: 134635 (C4 (Web Text))

Max Activating Token Index: 589

Text #19

Max Range: 0.7490. Min Range: -0.7490

Max Act: 0.4641. Min Act: -0.0000

Data Index: 883335 (C4 (Web Text))

Max Activating Token Index: 175