Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 77 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.6899. Min Act: -0.0001

Data Index: 1431885 (Python Code)

Max Activating Token Index: 686

Text #1

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.6461. Min Act: -0.0002

Data Index: 1128999 (C4 (Web Text))

Max Activating Token Index: 126

Text #2

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.6384. Min Act: -0.0002

Data Index: 14511 (C4 (Web Text))

Max Activating Token Index: 59

Text #3

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.6130. Min Act: -0.0001

Data Index: 825794 (C4 (Web Text))

Max Activating Token Index: 411

Text #4

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.6111. Min Act: -0.0001

Data Index: 1389248 (Python Code)

Max Activating Token Index: 146

Text #5

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5819. Min Act: -0.0001

Data Index: 1621139 (Python Code)

Max Activating Token Index: 657

Text #6

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5367. Min Act: -0.0002

Data Index: 1198334 (C4 (Web Text))

Max Activating Token Index: 389

Text #7

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5868. Min Act: -0.0002

Data Index: 183397 (C4 (Web Text))

Max Activating Token Index: 560

Text #8

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5777. Min Act: -0.0001

Data Index: 1529318 (Python Code)

Max Activating Token Index: 313

Text #9

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5790. Min Act: -0.0001

Data Index: 1390011 (Python Code)

Max Activating Token Index: 571

Text #10

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5594. Min Act: -0.0001

Data Index: 770498 (C4 (Web Text))

Max Activating Token Index: 265

Text #11

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5378. Min Act: -0.0002

Data Index: 977527 (C4 (Web Text))

Max Activating Token Index: 400

Text #12

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.4623. Min Act: -0.0001

Data Index: 1407784 (Python Code)

Max Activating Token Index: 776

Text #13

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.4993. Min Act: -0.0001

Data Index: 1455585 (Python Code)

Max Activating Token Index: 493

Text #14

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5420. Min Act: -0.0002

Data Index: 1174850 (C4 (Web Text))

Max Activating Token Index: 17

Text #15

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5390. Min Act: -0.0001

Data Index: 1334498 (C4 (Web Text))

Max Activating Token Index: 300

Text #16

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5229. Min Act: -0.0001

Data Index: 1510592 (Python Code)

Max Activating Token Index: 12

Text #17

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5208. Min Act: -0.0001

Data Index: 1589941 (Python Code)

Max Activating Token Index: 450

Text #18

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5020. Min Act: -0.0001

Data Index: 923369 (C4 (Web Text))

Max Activating Token Index: 334

Text #19

Max Range: 0.6899. Min Range: -0.6899

Max Act: 0.5151. Min Act: -0.0001

Data Index: 1432642 (Python Code)

Max Activating Token Index: 42