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

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

Neuron 4450 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: 4.7839. Min Range: -4.7839

Max Act: 4.7839. Min Act: -0.0000

Data Index: 321107 (C4 (Web Text))

Max Activating Token Index: 330

Text #1

Max Range: 4.7839. Min Range: -4.7839

Max Act: 4.4449. Min Act: -0.0000

Data Index: 676134 (C4 (Web Text))

Max Activating Token Index: 251

Text #2

Max Range: 4.7839. Min Range: -4.7839

Max Act: 4.1285. Min Act: -0.0000

Data Index: 371765 (C4 (Web Text))

Max Activating Token Index: 653

Text #3

Max Range: 4.7839. Min Range: -4.7839

Max Act: 4.2663. Min Act: -0.0000

Data Index: 850624 (C4 (Web Text))

Max Activating Token Index: 519

Text #4

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.9692. Min Act: -0.0000

Data Index: 612206 (C4 (Web Text))

Max Activating Token Index: 547

Text #5

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.9370. Min Act: -0.0000

Data Index: 150884 (C4 (Web Text))

Max Activating Token Index: 289

Text #6

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.9079. Min Act: -0.0000

Data Index: 732449 (C4 (Web Text))

Max Activating Token Index: 337

Text #7

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.7945. Min Act: -0.0000

Data Index: 1425485 (Python Code)

Max Activating Token Index: 477

Text #8

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.3311. Min Act: -0.0000

Data Index: 998404 (C4 (Web Text))

Max Activating Token Index: 929

Text #9

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.3477. Min Act: -0.0000

Data Index: 268658 (C4 (Web Text))

Max Activating Token Index: 242

Text #10

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.1740. Min Act: -0.0000

Data Index: 139581 (C4 (Web Text))

Max Activating Token Index: 162

Text #11

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.2728. Min Act: -0.0000

Data Index: 285043 (C4 (Web Text))

Max Activating Token Index: 583

Text #12

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.2531. Min Act: -0.0000

Data Index: 608128 (C4 (Web Text))

Max Activating Token Index: 233

Text #13

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.2574. Min Act: -0.0000

Data Index: 725983 (C4 (Web Text))

Max Activating Token Index: 39

Text #14

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.2286. Min Act: -0.0000

Data Index: 1612078 (Python Code)

Max Activating Token Index: 466

Text #15

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.2318. Min Act: -0.0000

Data Index: 969046 (C4 (Web Text))

Max Activating Token Index: 588

Text #16

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.3580. Min Act: -0.0000

Data Index: 407947 (C4 (Web Text))

Max Activating Token Index: 392

Text #17

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.0489. Min Act: -0.0000

Data Index: 986156 (C4 (Web Text))

Max Activating Token Index: 655

Text #18

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.2014. Min Act: -0.0000

Data Index: 1500987 (Python Code)

Max Activating Token Index: 881

Text #19

Max Range: 4.7839. Min Range: -4.7839

Max Act: 3.3531. Min Act: -0.0000

Data Index: 422638 (C4 (Web Text))

Max Activating Token Index: 832