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

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

Neuron 968 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1957. Min Range: -2.1957

Max Act: 2.1957. Min Act: -0.0000

Data Index: 293726 (C4 (Web Text))

Max Activating Token Index: 515

Text #1

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.6625. Min Act: -0.0000

Data Index: 1435953 (Python Code)

Max Activating Token Index: 931

Text #2

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.7614. Min Act: -0.0000

Data Index: 1098326 (C4 (Web Text))

Max Activating Token Index: 437

Text #3

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.5924. Min Act: -0.0000

Data Index: 1324288 (C4 (Web Text))

Max Activating Token Index: 778

Text #4

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.6441. Min Act: -0.0000

Data Index: 565314 (C4 (Web Text))

Max Activating Token Index: 76

Text #5

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.5189. Min Act: -0.0000

Data Index: 648990 (C4 (Web Text))

Max Activating Token Index: 708

Text #6

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.4735. Min Act: -0.0000

Data Index: 685087 (C4 (Web Text))

Max Activating Token Index: 775

Text #7

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.4814. Min Act: -0.0000

Data Index: 1144651 (C4 (Web Text))

Max Activating Token Index: 728

Text #8

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.5786. Min Act: -0.0000

Data Index: 49597 (C4 (Web Text))

Max Activating Token Index: 53

Text #9

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.5033. Min Act: -0.0000

Data Index: 430489 (C4 (Web Text))

Max Activating Token Index: 717

Text #10

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.5365. Min Act: -0.0000

Data Index: 1439867 (Python Code)

Max Activating Token Index: 627

Text #11

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.5117. Min Act: -0.0000

Data Index: 503206 (C4 (Web Text))

Max Activating Token Index: 34

Text #12

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.3259. Min Act: -0.0000

Data Index: 659968 (C4 (Web Text))

Max Activating Token Index: 839

Text #13

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.4115. Min Act: -0.0000

Data Index: 721428 (C4 (Web Text))

Max Activating Token Index: 311

Text #14

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.4052. Min Act: -0.0000

Data Index: 1250330 (C4 (Web Text))

Max Activating Token Index: 576

Text #15

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.3551. Min Act: -0.0000

Data Index: 773844 (C4 (Web Text))

Max Activating Token Index: 850

Text #16

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.3647. Min Act: -0.0000

Data Index: 612503 (C4 (Web Text))

Max Activating Token Index: 439

Text #17

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.3651. Min Act: -0.0000

Data Index: 1480664 (Python Code)

Max Activating Token Index: 88

Text #18

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.3249. Min Act: -0.0000

Data Index: 642620 (C4 (Web Text))

Max Activating Token Index: 517

Text #19

Max Range: 2.1957. Min Range: -2.1957

Max Act: 1.3976. Min Act: -0.0000

Data Index: 1118938 (C4 (Web Text))

Max Activating Token Index: 778