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

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

Neuron 2491 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7057. Min Range: -3.7057

Max Act: 3.7057. Min Act: -0.0000

Data Index: 1506892 (Python Code)

Max Activating Token Index: 430

Text #1

Max Range: 3.7057. Min Range: -3.7057

Max Act: 2.4586. Min Act: -0.0000

Data Index: 1472059 (Python Code)

Max Activating Token Index: 59

Text #2

Max Range: 3.7057. Min Range: -3.7057

Max Act: 2.2421. Min Act: -0.0000

Data Index: 1518297 (Python Code)

Max Activating Token Index: 790

Text #3

Max Range: 3.7057. Min Range: -3.7057

Max Act: 2.0139. Min Act: -0.0000

Data Index: 1399951 (Python Code)

Max Activating Token Index: 622

Text #4

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.6610. Min Act: -0.0000

Data Index: 1478570 (Python Code)

Max Activating Token Index: 146

Text #5

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.8533. Min Act: -0.0000

Data Index: 1572391 (Python Code)

Max Activating Token Index: 864

Text #6

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.7322. Min Act: -0.0000

Data Index: 426474 (C4 (Web Text))

Max Activating Token Index: 716

Text #7

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.6632. Min Act: -0.0000

Data Index: 1452200 (Python Code)

Max Activating Token Index: 520

Text #8

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.6419. Min Act: -0.0000

Data Index: 659591 (C4 (Web Text))

Max Activating Token Index: 219

Text #9

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.6310. Min Act: -0.0000

Data Index: 1460275 (Python Code)

Max Activating Token Index: 878

Text #10

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.5029. Min Act: -0.0000

Data Index: 1260396 (C4 (Web Text))

Max Activating Token Index: 504

Text #11

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.5844. Min Act: -0.0000

Data Index: 1496137 (Python Code)

Max Activating Token Index: 13

Text #12

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.4640. Min Act: -0.0000

Data Index: 133963 (C4 (Web Text))

Max Activating Token Index: 314

Text #13

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.6186. Min Act: -0.0000

Data Index: 1643310 (Python Code)

Max Activating Token Index: 331

Text #14

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.5097. Min Act: -0.0000

Data Index: 596236 (C4 (Web Text))

Max Activating Token Index: 407

Text #15

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.3389. Min Act: -0.0000

Data Index: 1465743 (Python Code)

Max Activating Token Index: 326

Text #16

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.3797. Min Act: -0.0000

Data Index: 1451941 (Python Code)

Max Activating Token Index: 95

Text #17

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.4706. Min Act: -0.0000

Data Index: 1537718 (Python Code)

Max Activating Token Index: 620

Text #18

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.3387. Min Act: -0.0000

Data Index: 498083 (C4 (Web Text))

Max Activating Token Index: 496

Text #19

Max Range: 3.7057. Min Range: -3.7057

Max Act: 1.3311. Min Act: -0.0000

Data Index: 1531326 (Python Code)

Max Activating Token Index: 772