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

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

Neuron 4479 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: 0.4064. Min Range: -0.4064

Max Act: 0.4064. Min Act: -0.0000

Data Index: 80658 (C4 (Web Text))

Max Activating Token Index: 572

Text #1

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.3429. Min Act: -0.0000

Data Index: 21262 (C4 (Web Text))

Max Activating Token Index: 900

Text #2

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.3406. Min Act: -0.0000

Data Index: 867310 (C4 (Web Text))

Max Activating Token Index: 592

Text #3

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.3523. Min Act: -0.0000

Data Index: 290893 (C4 (Web Text))

Max Activating Token Index: 482

Text #4

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.3239. Min Act: -0.0000

Data Index: 1099390 (C4 (Web Text))

Max Activating Token Index: 966

Text #5

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.3180. Min Act: -0.0000

Data Index: 1023411 (C4 (Web Text))

Max Activating Token Index: 619

Text #6

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2878. Min Act: -0.0000

Data Index: 1040361 (C4 (Web Text))

Max Activating Token Index: 759

Text #7

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2821. Min Act: -0.0000

Data Index: 150129 (C4 (Web Text))

Max Activating Token Index: 855

Text #8

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.3009. Min Act: -0.0000

Data Index: 529315 (C4 (Web Text))

Max Activating Token Index: 559

Text #9

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.3037. Min Act: -0.0000

Data Index: 980169 (C4 (Web Text))

Max Activating Token Index: 480

Text #10

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2771. Min Act: -0.0000

Data Index: 536681 (C4 (Web Text))

Max Activating Token Index: 76

Text #11

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2908. Min Act: -0.0000

Data Index: 1341586 (C4 (Web Text))

Max Activating Token Index: 1021

Text #12

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2630. Min Act: -0.0000

Data Index: 627533 (C4 (Web Text))

Max Activating Token Index: 570

Text #13

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2662. Min Act: -0.0000

Data Index: 575244 (C4 (Web Text))

Max Activating Token Index: 382

Text #14

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2587. Min Act: -0.0000

Data Index: 1608808 (Python Code)

Max Activating Token Index: 332

Text #15

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2683. Min Act: -0.0000

Data Index: 869635 (C4 (Web Text))

Max Activating Token Index: 549

Text #16

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2516. Min Act: -0.0000

Data Index: 543607 (C4 (Web Text))

Max Activating Token Index: 479

Text #17

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2751. Min Act: -0.0000

Data Index: 578782 (C4 (Web Text))

Max Activating Token Index: 234

Text #18

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2491. Min Act: -0.0000

Data Index: 415040 (C4 (Web Text))

Max Activating Token Index: 510

Text #19

Max Range: 0.4064. Min Range: -0.4064

Max Act: 0.2564. Min Act: -0.0000

Data Index: 222889 (C4 (Web Text))

Max Activating Token Index: 999