Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1532 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3522. Min Act: -0.0001

Data Index: 1257744 (C4 (Web Text))

Max Activating Token Index: 977

Text #1

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.4311. Min Act: -0.0001

Data Index: 535177 (C4 (Web Text))

Max Activating Token Index: 182

Text #2

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3590. Min Act: -0.0001

Data Index: 857531 (C4 (Web Text))

Max Activating Token Index: 524

Text #3

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3625. Min Act: -0.0001

Data Index: 298241 (C4 (Web Text))

Max Activating Token Index: 731

Text #4

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3422. Min Act: -0.0001

Data Index: 295931 (C4 (Web Text))

Max Activating Token Index: 67

Text #5

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3684. Min Act: -0.0001

Data Index: 1402583 (Python Code)

Max Activating Token Index: 778

Text #6

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.4530. Min Act: -0.0001

Data Index: 1055078 (C4 (Web Text))

Max Activating Token Index: 274

Text #7

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3544. Min Act: -0.0001

Data Index: 1267130 (C4 (Web Text))

Max Activating Token Index: 644

Text #8

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3558. Min Act: -0.0001

Data Index: 1015132 (C4 (Web Text))

Max Activating Token Index: 332

Text #9

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3726. Min Act: -0.0001

Data Index: 930577 (C4 (Web Text))

Max Activating Token Index: 830

Text #10

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3604. Min Act: -0.0001

Data Index: 787562 (C4 (Web Text))

Max Activating Token Index: 128

Text #11

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3352. Min Act: -0.0001

Data Index: 1497623 (Python Code)

Max Activating Token Index: 89

Text #12

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3452. Min Act: -0.0001

Data Index: 1579109 (Python Code)

Max Activating Token Index: 404

Text #13

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3436. Min Act: -0.0001

Data Index: 189571 (C4 (Web Text))

Max Activating Token Index: 731

Text #14

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3549. Min Act: -0.0001

Data Index: 1375127 (Python Code)

Max Activating Token Index: 743

Text #15

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3057. Min Act: -0.0001

Data Index: 254402 (C4 (Web Text))

Max Activating Token Index: 809

Text #16

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3537. Min Act: -0.0001

Data Index: 201487 (C4 (Web Text))

Max Activating Token Index: 107

Text #17

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.2937. Min Act: -0.0001

Data Index: 922541 (C4 (Web Text))

Max Activating Token Index: 999

Text #18

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3400. Min Act: -0.0001

Data Index: 451848 (C4 (Web Text))

Max Activating Token Index: 789

Text #19

Max Range: 0.4530. Min Range: -0.4530

Max Act: 0.3754. Min Act: -0.0001

Data Index: 914719 (C4 (Web Text))

Max Activating Token Index: 832