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

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

Neuron 3510 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: 1.3309. Min Range: -1.3309

Max Act: 1.3309. Min Act: -0.0000

Data Index: 166768 (C4 (Web Text))

Max Activating Token Index: 401

Text #1

Max Range: 1.3309. Min Range: -1.3309

Max Act: 1.1819. Min Act: -0.0000

Data Index: 829022 (C4 (Web Text))

Max Activating Token Index: 488

Text #2

Max Range: 1.3309. Min Range: -1.3309

Max Act: 1.0959. Min Act: -0.0000

Data Index: 429837 (C4 (Web Text))

Max Activating Token Index: 557

Text #3

Max Range: 1.3309. Min Range: -1.3309

Max Act: 1.1427. Min Act: -0.0000

Data Index: 586815 (C4 (Web Text))

Max Activating Token Index: 566

Text #4

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.9803. Min Act: -0.0000

Data Index: 649781 (C4 (Web Text))

Max Activating Token Index: 374

Text #5

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.8913. Min Act: -0.0000

Data Index: 822391 (C4 (Web Text))

Max Activating Token Index: 765

Text #6

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.8830. Min Act: -0.0000

Data Index: 1502821 (Python Code)

Max Activating Token Index: 474

Text #7

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.8913. Min Act: -0.0000

Data Index: 532040 (C4 (Web Text))

Max Activating Token Index: 515

Text #8

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.9576. Min Act: -0.0000

Data Index: 313064 (C4 (Web Text))

Max Activating Token Index: 158

Text #9

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.8529. Min Act: -0.0000

Data Index: 20436 (C4 (Web Text))

Max Activating Token Index: 322

Text #10

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.8124. Min Act: -0.0000

Data Index: 1562773 (Python Code)

Max Activating Token Index: 810

Text #11

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.7722. Min Act: -0.0000

Data Index: 1526002 (Python Code)

Max Activating Token Index: 857

Text #12

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.8067. Min Act: -0.0000

Data Index: 445326 (C4 (Web Text))

Max Activating Token Index: 459

Text #13

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.7828. Min Act: -0.0000

Data Index: 1636623 (Python Code)

Max Activating Token Index: 702

Text #14

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.8076. Min Act: -0.0000

Data Index: 984981 (C4 (Web Text))

Max Activating Token Index: 558

Text #15

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.7847. Min Act: -0.0000

Data Index: 343588 (C4 (Web Text))

Max Activating Token Index: 193

Text #16

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.7923. Min Act: -0.0000

Data Index: 201603 (C4 (Web Text))

Max Activating Token Index: 133

Text #17

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.7822. Min Act: -0.0000

Data Index: 1534545 (Python Code)

Max Activating Token Index: 510

Text #18

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.7499. Min Act: -0.0000

Data Index: 510875 (C4 (Web Text))

Max Activating Token Index: 563

Text #19

Max Range: 1.3309. Min Range: -1.3309

Max Act: 0.7575. Min Act: -0.0000

Data Index: 1168356 (C4 (Web Text))

Max Activating Token Index: 921