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

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

Neuron 2925 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.6931. Min Act: 0.0000

Data Index: 743143 (C4 (Web Text))

Max Activating Token Index: 167

Text #1

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.6938. Min Act: -0.0000

Data Index: 222213 (C4 (Web Text))

Max Activating Token Index: 673

Text #2

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.6435. Min Act: -0.0000

Data Index: 215706 (C4 (Web Text))

Max Activating Token Index: 889

Text #3

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.6357. Min Act: -0.0000

Data Index: 1411797 (Python Code)

Max Activating Token Index: 856

Text #4

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5956. Min Act: -0.0000

Data Index: 549547 (C4 (Web Text))

Max Activating Token Index: 859

Text #5

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.6085. Min Act: 0.0000

Data Index: 360869 (C4 (Web Text))

Max Activating Token Index: 676

Text #6

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5933. Min Act: -0.0000

Data Index: 1452132 (Python Code)

Max Activating Token Index: 312

Text #7

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5837. Min Act: 0.0000

Data Index: 1558699 (Python Code)

Max Activating Token Index: 709

Text #8

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5785. Min Act: -0.0000

Data Index: 1402138 (Python Code)

Max Activating Token Index: 497

Text #9

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5669. Min Act: -0.0000

Data Index: 1061429 (C4 (Web Text))

Max Activating Token Index: 735

Text #10

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5259. Min Act: -0.0000

Data Index: 1479582 (Python Code)

Max Activating Token Index: 671

Text #11

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5768. Min Act: 0.0000

Data Index: 1236720 (C4 (Web Text))

Max Activating Token Index: 642

Text #12

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.4817. Min Act: -0.0000

Data Index: 340591 (C4 (Web Text))

Max Activating Token Index: 967

Text #13

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5650. Min Act: -0.0000

Data Index: 567813 (C4 (Web Text))

Max Activating Token Index: 390

Text #14

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5558. Min Act: 0.0000

Data Index: 1360637 (Python Code)

Max Activating Token Index: 632

Text #15

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5355. Min Act: -0.0000

Data Index: 429643 (C4 (Web Text))

Max Activating Token Index: 642

Text #16

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5098. Min Act: -0.0000

Data Index: 1072625 (C4 (Web Text))

Max Activating Token Index: 570

Text #17

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5030. Min Act: -0.0000

Data Index: 377361 (C4 (Web Text))

Max Activating Token Index: 555

Text #18

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5016. Min Act: -0.0000

Data Index: 1438901 (Python Code)

Max Activating Token Index: 814

Text #19

Max Range: 0.6938. Min Range: -0.6938

Max Act: 0.5115. Min Act: -0.0000

Data Index: 8795 (C4 (Web Text))

Max Activating Token Index: 397