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

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

Neuron 97 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: 4.0826. Min Range: -4.0826

Max Act: 4.0826. Min Act: -0.0000

Data Index: 190555 (C4 (Web Text))

Max Activating Token Index: 878

Text #1

Max Range: 4.0826. Min Range: -4.0826

Max Act: 3.6506. Min Act: -0.0000

Data Index: 1579805 (Python Code)

Max Activating Token Index: 407

Text #2

Max Range: 4.0826. Min Range: -4.0826

Max Act: 3.2828. Min Act: -0.0000

Data Index: 23419 (C4 (Web Text))

Max Activating Token Index: 671

Text #3

Max Range: 4.0826. Min Range: -4.0826

Max Act: 3.3423. Min Act: 0.0000

Data Index: 118440 (C4 (Web Text))

Max Activating Token Index: 869

Text #4

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.7378. Min Act: -0.0000

Data Index: 1600717 (Python Code)

Max Activating Token Index: 866

Text #5

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.7902. Min Act: -0.0000

Data Index: 1414919 (Python Code)

Max Activating Token Index: 794

Text #6

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.6114. Min Act: 0.0000

Data Index: 1318751 (C4 (Web Text))

Max Activating Token Index: 838

Text #7

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.6493. Min Act: -0.0000

Data Index: 1026857 (C4 (Web Text))

Max Activating Token Index: 807

Text #8

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.6023. Min Act: -0.0000

Data Index: 1353493 (C4 (Web Text))

Max Activating Token Index: 989

Text #9

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.5643. Min Act: -0.0000

Data Index: 1439025 (Python Code)

Max Activating Token Index: 1005

Text #10

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.4575. Min Act: -0.0000

Data Index: 1438772 (Python Code)

Max Activating Token Index: 732

Text #11

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.7783. Min Act: -0.0000

Data Index: 589931 (C4 (Web Text))

Max Activating Token Index: 878

Text #12

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.4194. Min Act: 0.0000

Data Index: 949397 (C4 (Web Text))

Max Activating Token Index: 486

Text #13

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.4726. Min Act: -0.0000

Data Index: 1577472 (Python Code)

Max Activating Token Index: 983

Text #14

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.3734. Min Act: -0.0000

Data Index: 1019139 (C4 (Web Text))

Max Activating Token Index: 353

Text #15

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.3578. Min Act: -0.0000

Data Index: 752110 (C4 (Web Text))

Max Activating Token Index: 523

Text #16

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.3412. Min Act: -0.0000

Data Index: 216121 (C4 (Web Text))

Max Activating Token Index: 615

Text #17

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.5080. Min Act: -0.0000

Data Index: 1553980 (Python Code)

Max Activating Token Index: 680

Text #18

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.2386. Min Act: -0.0000

Data Index: 1285280 (C4 (Web Text))

Max Activating Token Index: 535

Text #19

Max Range: 4.0826. Min Range: -4.0826

Max Act: 2.3654. Min Act: -0.0000

Data Index: 531080 (C4 (Web Text))

Max Activating Token Index: 276