Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2392 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.9526. Min Act: -0.0001

Data Index: 1065511 (C4 (Web Text))

Max Activating Token Index: 847

Text #1

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.9055. Min Act: -0.0001

Data Index: 938824 (C4 (Web Text))

Max Activating Token Index: 949

Text #2

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.7346. Min Act: -0.0001

Data Index: 139685 (C4 (Web Text))

Max Activating Token Index: 961

Text #3

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.6896. Min Act: -0.0001

Data Index: 1338951 (C4 (Web Text))

Max Activating Token Index: 752

Text #4

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.7006. Min Act: -0.0001

Data Index: 1356753 (C4 (Web Text))

Max Activating Token Index: 538

Text #5

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.6649. Min Act: -0.0001

Data Index: 1525774 (Python Code)

Max Activating Token Index: 980

Text #6

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.6104. Min Act: -0.0001

Data Index: 1399583 (Python Code)

Max Activating Token Index: 922

Text #7

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.6565. Min Act: -0.0001

Data Index: 291132 (C4 (Web Text))

Max Activating Token Index: 251

Text #8

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.6693. Min Act: -0.0001

Data Index: 1464665 (Python Code)

Max Activating Token Index: 321

Text #9

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.6331. Min Act: -0.0001

Data Index: 170476 (C4 (Web Text))

Max Activating Token Index: 330

Text #10

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.5510. Min Act: -0.0001

Data Index: 1322180 (C4 (Web Text))

Max Activating Token Index: 39

Text #11

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.5987. Min Act: -0.0001

Data Index: 1585196 (Python Code)

Max Activating Token Index: 101

Text #12

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.6967. Min Act: -0.0001

Data Index: 1198710 (C4 (Web Text))

Max Activating Token Index: 589

Text #13

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.6086. Min Act: -0.0001

Data Index: 1455585 (Python Code)

Max Activating Token Index: 175

Text #14

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.6411. Min Act: -0.0001

Data Index: 315144 (C4 (Web Text))

Max Activating Token Index: 286

Text #15

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.5840. Min Act: -0.0001

Data Index: 1578622 (Python Code)

Max Activating Token Index: 707

Text #16

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.5616. Min Act: -0.0001

Data Index: 1487458 (Python Code)

Max Activating Token Index: 768

Text #17

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.5892. Min Act: -0.0001

Data Index: 1088990 (C4 (Web Text))

Max Activating Token Index: 869

Text #18

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.5754. Min Act: -0.0001

Data Index: 153639 (C4 (Web Text))

Max Activating Token Index: 15

Text #19

Max Range: 0.9526. Min Range: -0.9526

Max Act: 0.5557. Min Act: -0.0001

Data Index: 1484885 (Python Code)

Max Activating Token Index: 975