Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5821 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.3717. Min Range: -1.3717

Max Act: 1.3640. Min Act: -0.0000

Data Index: 1063491 (C4 (Web Text))

Max Activating Token Index: 30

Text #1

Max Range: 1.3717. Min Range: -1.3717

Max Act: 1.3717. Min Act: -0.0000

Data Index: 1141607 (C4 (Web Text))

Max Activating Token Index: 786

Text #2

Max Range: 1.3717. Min Range: -1.3717

Max Act: 1.3702. Min Act: -0.0000

Data Index: 209109 (C4 (Web Text))

Max Activating Token Index: 864

Text #3

Max Range: 1.3717. Min Range: -1.3717

Max Act: 1.1393. Min Act: 0.0000

Data Index: 1445698 (Python Code)

Max Activating Token Index: 450

Text #4

Max Range: 1.3717. Min Range: -1.3717

Max Act: 1.1898. Min Act: -0.0000

Data Index: 1362727 (Python Code)

Max Activating Token Index: 901

Text #5

Max Range: 1.3717. Min Range: -1.3717

Max Act: 1.2518. Min Act: -0.0000

Data Index: 818607 (C4 (Web Text))

Max Activating Token Index: 1021

Text #6

Max Range: 1.3717. Min Range: -1.3717

Max Act: 1.2115. Min Act: -0.0000

Data Index: 877121 (C4 (Web Text))

Max Activating Token Index: 39

Text #7

Max Range: 1.3717. Min Range: -1.3717

Max Act: 1.0578. Min Act: -0.0000

Data Index: 878889 (C4 (Web Text))

Max Activating Token Index: 405

Text #8

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.9776. Min Act: -0.0000

Data Index: 1655057 (Python Code)

Max Activating Token Index: 604

Text #9

Max Range: 1.3717. Min Range: -1.3717

Max Act: 1.0218. Min Act: -0.0000

Data Index: 1149405 (C4 (Web Text))

Max Activating Token Index: 497

Text #10

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.8834. Min Act: -0.0000

Data Index: 1623757 (Python Code)

Max Activating Token Index: 898

Text #11

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.9133. Min Act: 0.0000

Data Index: 1465430 (Python Code)

Max Activating Token Index: 611

Text #12

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.8530. Min Act: -0.0000

Data Index: 1528631 (Python Code)

Max Activating Token Index: 860

Text #13

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.8817. Min Act: -0.0000

Data Index: 594372 (C4 (Web Text))

Max Activating Token Index: 305

Text #14

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.8188. Min Act: -0.0000

Data Index: 1643505 (Python Code)

Max Activating Token Index: 61

Text #15

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.8885. Min Act: -0.0000

Data Index: 594752 (C4 (Web Text))

Max Activating Token Index: 920

Text #16

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.7488. Min Act: -0.0000

Data Index: 1377846 (Python Code)

Max Activating Token Index: 355

Text #17

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.7830. Min Act: 0.0000

Data Index: 1486788 (Python Code)

Max Activating Token Index: 615

Text #18

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.7221. Min Act: -0.0000

Data Index: 923776 (C4 (Web Text))

Max Activating Token Index: 827

Text #19

Max Range: 1.3717. Min Range: -1.3717

Max Act: 0.7141. Min Act: -0.0000

Data Index: 1461455 (Python Code)

Max Activating Token Index: 674