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

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

Neuron 3714 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: 0.4759. Min Range: -0.4759

Max Act: 0.4759. Min Act: -0.0000

Data Index: 629752 (C4 (Web Text))

Max Activating Token Index: 353

Text #1

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.3368. Min Act: -0.0000

Data Index: 966634 (C4 (Web Text))

Max Activating Token Index: 317

Text #2

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.3394. Min Act: -0.0000

Data Index: 1388744 (Python Code)

Max Activating Token Index: 63

Text #3

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.3117. Min Act: -0.0000

Data Index: 727501 (C4 (Web Text))

Max Activating Token Index: 822

Text #4

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.3134. Min Act: -0.0000

Data Index: 1272676 (C4 (Web Text))

Max Activating Token Index: 302

Text #5

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.3038. Min Act: -0.0000

Data Index: 973384 (C4 (Web Text))

Max Activating Token Index: 568

Text #6

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.3116. Min Act: -0.0000

Data Index: 1023815 (C4 (Web Text))

Max Activating Token Index: 497

Text #7

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2977. Min Act: -0.0000

Data Index: 745987 (C4 (Web Text))

Max Activating Token Index: 761

Text #8

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2773. Min Act: -0.0000

Data Index: 1405173 (Python Code)

Max Activating Token Index: 36

Text #9

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2920. Min Act: -0.0000

Data Index: 1467345 (Python Code)

Max Activating Token Index: 106

Text #10

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2670. Min Act: -0.0000

Data Index: 1596318 (Python Code)

Max Activating Token Index: 122

Text #11

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2853. Min Act: -0.0000

Data Index: 603237 (C4 (Web Text))

Max Activating Token Index: 535

Text #12

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2879. Min Act: -0.0000

Data Index: 900173 (C4 (Web Text))

Max Activating Token Index: 782

Text #13

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2615. Min Act: -0.0000

Data Index: 296821 (C4 (Web Text))

Max Activating Token Index: 633

Text #14

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2818. Min Act: -0.0000

Data Index: 294558 (C4 (Web Text))

Max Activating Token Index: 665

Text #15

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2726. Min Act: -0.0000

Data Index: 824601 (C4 (Web Text))

Max Activating Token Index: 637

Text #16

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2679. Min Act: -0.0000

Data Index: 1530578 (Python Code)

Max Activating Token Index: 42

Text #17

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2532. Min Act: -0.0000

Data Index: 1004784 (C4 (Web Text))

Max Activating Token Index: 210

Text #18

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2736. Min Act: -0.0000

Data Index: 410117 (C4 (Web Text))

Max Activating Token Index: 784

Text #19

Max Range: 0.4759. Min Range: -0.4759

Max Act: 0.2593. Min Act: -0.0000

Data Index: 650487 (C4 (Web Text))

Max Activating Token Index: 364