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

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

Neuron 3725 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4883. Min Range: -1.4883

Max Act: 1.4883. Min Act: -0.0000

Data Index: 585222 (C4 (Web Text))

Max Activating Token Index: 471

Text #1

Max Range: 1.4883. Min Range: -1.4883

Max Act: 1.2265. Min Act: -0.0001

Data Index: 504201 (C4 (Web Text))

Max Activating Token Index: 861

Text #2

Max Range: 1.4883. Min Range: -1.4883

Max Act: 1.2687. Min Act: -0.0001

Data Index: 1027876 (C4 (Web Text))

Max Activating Token Index: 738

Text #3

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.9924. Min Act: -0.0001

Data Index: 398152 (C4 (Web Text))

Max Activating Token Index: 396

Text #4

Max Range: 1.4883. Min Range: -1.4883

Max Act: 1.0652. Min Act: -0.0000

Data Index: 1537575 (Python Code)

Max Activating Token Index: 760

Text #5

Max Range: 1.4883. Min Range: -1.4883

Max Act: 1.0078. Min Act: -0.0000

Data Index: 1334987 (C4 (Web Text))

Max Activating Token Index: 56

Text #6

Max Range: 1.4883. Min Range: -1.4883

Max Act: 1.0679. Min Act: -0.0000

Data Index: 846699 (C4 (Web Text))

Max Activating Token Index: 331

Text #7

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.9833. Min Act: -0.0000

Data Index: 227833 (C4 (Web Text))

Max Activating Token Index: 291

Text #8

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.9587. Min Act: -0.0000

Data Index: 232988 (C4 (Web Text))

Max Activating Token Index: 175

Text #9

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.9946. Min Act: -0.0001

Data Index: 1089209 (C4 (Web Text))

Max Activating Token Index: 62

Text #10

Max Range: 1.4883. Min Range: -1.4883

Max Act: 1.0217. Min Act: -0.0001

Data Index: 485675 (C4 (Web Text))

Max Activating Token Index: 453

Text #11

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.9202. Min Act: -0.0001

Data Index: 869533 (C4 (Web Text))

Max Activating Token Index: 581

Text #12

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.8635. Min Act: -0.0000

Data Index: 1119875 (C4 (Web Text))

Max Activating Token Index: 39

Text #13

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.9421. Min Act: -0.0000

Data Index: 176774 (C4 (Web Text))

Max Activating Token Index: 779

Text #14

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.8424. Min Act: -0.0000

Data Index: 637634 (C4 (Web Text))

Max Activating Token Index: 527

Text #15

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.9289. Min Act: -0.0000

Data Index: 704173 (C4 (Web Text))

Max Activating Token Index: 675

Text #16

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.9176. Min Act: -0.0000

Data Index: 736695 (C4 (Web Text))

Max Activating Token Index: 94

Text #17

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.8593. Min Act: -0.0000

Data Index: 34434 (C4 (Web Text))

Max Activating Token Index: 207

Text #18

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.8153. Min Act: -0.0000

Data Index: 983691 (C4 (Web Text))

Max Activating Token Index: 800

Text #19

Max Range: 1.4883. Min Range: -1.4883

Max Act: 0.8482. Min Act: -0.0000

Data Index: 1165029 (C4 (Web Text))

Max Activating Token Index: 863