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

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

Neuron 1108 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2440. Min Act: -0.0000

Data Index: 1402619 (Python Code)

Max Activating Token Index: 539

Text #1

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2485. Min Act: -0.0000

Data Index: 1393883 (Python Code)

Max Activating Token Index: 562

Text #2

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2479. Min Act: -0.0000

Data Index: 1378432 (Python Code)

Max Activating Token Index: 844

Text #3

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2445. Min Act: -0.0000

Data Index: 1552033 (Python Code)

Max Activating Token Index: 131

Text #4

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2388. Min Act: -0.0000

Data Index: 1002534 (C4 (Web Text))

Max Activating Token Index: 647

Text #5

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2240. Min Act: -0.0000

Data Index: 1640423 (Python Code)

Max Activating Token Index: 177

Text #6

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2212. Min Act: -0.0000

Data Index: 120483 (C4 (Web Text))

Max Activating Token Index: 490

Text #7

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2127. Min Act: -0.0000

Data Index: 880635 (C4 (Web Text))

Max Activating Token Index: 350

Text #8

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2135. Min Act: -0.0000

Data Index: 136304 (C4 (Web Text))

Max Activating Token Index: 296

Text #9

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2052. Min Act: -0.0000

Data Index: 480684 (C4 (Web Text))

Max Activating Token Index: 376

Text #10

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2077. Min Act: -0.0000

Data Index: 600473 (C4 (Web Text))

Max Activating Token Index: 683

Text #11

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2003. Min Act: -0.0000

Data Index: 341131 (C4 (Web Text))

Max Activating Token Index: 532

Text #12

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2042. Min Act: -0.0000

Data Index: 1620591 (Python Code)

Max Activating Token Index: 874

Text #13

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2071. Min Act: -0.0000

Data Index: 380161 (C4 (Web Text))

Max Activating Token Index: 95

Text #14

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.1874. Min Act: -0.0000

Data Index: 867483 (C4 (Web Text))

Max Activating Token Index: 565

Text #15

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.1935. Min Act: -0.0000

Data Index: 7747 (C4 (Web Text))

Max Activating Token Index: 833

Text #16

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2049. Min Act: -0.0000

Data Index: 858762 (C4 (Web Text))

Max Activating Token Index: 114

Text #17

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.1903. Min Act: -0.0000

Data Index: 1109118 (C4 (Web Text))

Max Activating Token Index: 712

Text #18

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2120. Min Act: -0.0000

Data Index: 662652 (C4 (Web Text))

Max Activating Token Index: 656

Text #19

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2009. Min Act: -0.0000

Data Index: 251638 (C4 (Web Text))

Max Activating Token Index: 449