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

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

Neuron 3077 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.6778. Min Act: -0.0000

Data Index: 1608702 (Python Code)

Max Activating Token Index: 712

Text #1

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.6978. Min Act: -0.0000

Data Index: 332022 (C4 (Web Text))

Max Activating Token Index: 355

Text #2

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.6233. Min Act: -0.0000

Data Index: 776582 (C4 (Web Text))

Max Activating Token Index: 631

Text #3

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.5918. Min Act: -0.0000

Data Index: 1439876 (Python Code)

Max Activating Token Index: 305

Text #4

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.5489. Min Act: -0.0000

Data Index: 1622182 (Python Code)

Max Activating Token Index: 726

Text #5

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.5421. Min Act: -0.0000

Data Index: 1094132 (C4 (Web Text))

Max Activating Token Index: 744

Text #6

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.5366. Min Act: -0.0000

Data Index: 831216 (C4 (Web Text))

Max Activating Token Index: 741

Text #7

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.5828. Min Act: -0.0000

Data Index: 1021236 (C4 (Web Text))

Max Activating Token Index: 905

Text #8

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.5268. Min Act: -0.0000

Data Index: 622345 (C4 (Web Text))

Max Activating Token Index: 521

Text #9

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4878. Min Act: -0.0000

Data Index: 887542 (C4 (Web Text))

Max Activating Token Index: 542

Text #10

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4510. Min Act: -0.0000

Data Index: 288747 (C4 (Web Text))

Max Activating Token Index: 603

Text #11

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.5239. Min Act: -0.0000

Data Index: 1151783 (C4 (Web Text))

Max Activating Token Index: 808

Text #12

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4808. Min Act: -0.0000

Data Index: 122409 (C4 (Web Text))

Max Activating Token Index: 599

Text #13

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4815. Min Act: -0.0000

Data Index: 311592 (C4 (Web Text))

Max Activating Token Index: 422

Text #14

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4741. Min Act: -0.0000

Data Index: 921353 (C4 (Web Text))

Max Activating Token Index: 89

Text #15

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4633. Min Act: -0.0000

Data Index: 1613779 (Python Code)

Max Activating Token Index: 69

Text #16

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4454. Min Act: -0.0000

Data Index: 596832 (C4 (Web Text))

Max Activating Token Index: 999

Text #17

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4601. Min Act: -0.0000

Data Index: 1356903 (C4 (Web Text))

Max Activating Token Index: 779

Text #18

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4865. Min Act: -0.0000

Data Index: 577966 (C4 (Web Text))

Max Activating Token Index: 531

Text #19

Max Range: 0.6978. Min Range: -0.6978

Max Act: 0.4474. Min Act: -0.0000

Data Index: 1613596 (Python Code)

Max Activating Token Index: 589