Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 55 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.2482. Min Act: -0.0001

Data Index: 696402 (C4 (Web Text))

Max Activating Token Index: 764

Text #1

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.2319. Min Act: -0.0001

Data Index: 164926 (C4 (Web Text))

Max Activating Token Index: 132

Text #2

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1875. Min Act: -0.0001

Data Index: 18609 (C4 (Web Text))

Max Activating Token Index: 98

Text #3

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1795. Min Act: -0.0000

Data Index: 1097112 (C4 (Web Text))

Max Activating Token Index: 828

Text #4

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1781. Min Act: -0.0000

Data Index: 663814 (C4 (Web Text))

Max Activating Token Index: 166

Text #5

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1687. Min Act: -0.0001

Data Index: 908057 (C4 (Web Text))

Max Activating Token Index: 583

Text #6

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1686. Min Act: -0.0001

Data Index: 1342592 (C4 (Web Text))

Max Activating Token Index: 406

Text #7

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1745. Min Act: -0.0000

Data Index: 216031 (C4 (Web Text))

Max Activating Token Index: 585

Text #8

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1604. Min Act: -0.0001

Data Index: 719711 (C4 (Web Text))

Max Activating Token Index: 958

Text #9

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1609. Min Act: -0.0000

Data Index: 30792 (C4 (Web Text))

Max Activating Token Index: 953

Text #10

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1677. Min Act: -0.0001

Data Index: 103296 (C4 (Web Text))

Max Activating Token Index: 718

Text #11

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1607. Min Act: -0.0001

Data Index: 93486 (C4 (Web Text))

Max Activating Token Index: 456

Text #12

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1612. Min Act: -0.0001

Data Index: 734784 (C4 (Web Text))

Max Activating Token Index: 385

Text #13

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1692. Min Act: -0.0001

Data Index: 283302 (C4 (Web Text))

Max Activating Token Index: 445

Text #14

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1585. Min Act: -0.0000

Data Index: 164237 (C4 (Web Text))

Max Activating Token Index: 850

Text #15

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1533. Min Act: -0.0000

Data Index: 1109931 (C4 (Web Text))

Max Activating Token Index: 63

Text #16

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1564. Min Act: -0.0000

Data Index: 758361 (C4 (Web Text))

Max Activating Token Index: 315

Text #17

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1571. Min Act: -0.0001

Data Index: 89840 (C4 (Web Text))

Max Activating Token Index: 572

Text #18

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1454. Min Act: -0.0001

Data Index: 706469 (C4 (Web Text))

Max Activating Token Index: 1001

Text #19

Max Range: 0.2482. Min Range: -0.2482

Max Act: 0.1477. Min Act: -0.0000

Data Index: 699942 (C4 (Web Text))

Max Activating Token Index: 428