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

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

Neuron 4822 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1494. Min Range: -3.1494

Max Act: 3.1494. Min Act: -0.0000

Data Index: 320358 (C4 (Web Text))

Max Activating Token Index: 786

Text #1

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.7595. Min Act: -0.0000

Data Index: 324508 (C4 (Web Text))

Max Activating Token Index: 148

Text #2

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.7153. Min Act: -0.0000

Data Index: 621816 (C4 (Web Text))

Max Activating Token Index: 427

Text #3

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.5061. Min Act: -0.0000

Data Index: 908434 (C4 (Web Text))

Max Activating Token Index: 553

Text #4

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.5744. Min Act: -0.0000

Data Index: 1835 (C4 (Web Text))

Max Activating Token Index: 960

Text #5

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.3924. Min Act: -0.0000

Data Index: 789017 (C4 (Web Text))

Max Activating Token Index: 502

Text #6

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.2812. Min Act: -0.0000

Data Index: 1239709 (C4 (Web Text))

Max Activating Token Index: 63

Text #7

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.2503. Min Act: -0.0000

Data Index: 58049 (C4 (Web Text))

Max Activating Token Index: 169

Text #8

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.0673. Min Act: -0.0000

Data Index: 1554866 (Python Code)

Max Activating Token Index: 1015

Text #9

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.0388. Min Act: -0.0000

Data Index: 1445038 (Python Code)

Max Activating Token Index: 600

Text #10

Max Range: 3.1494. Min Range: -3.1494

Max Act: 1.9673. Min Act: -0.0000

Data Index: 313862 (C4 (Web Text))

Max Activating Token Index: 962

Text #11

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.0720. Min Act: -0.0000

Data Index: 49499 (C4 (Web Text))

Max Activating Token Index: 405

Text #12

Max Range: 3.1494. Min Range: -3.1494

Max Act: 2.0419. Min Act: -0.0000

Data Index: 1623334 (Python Code)

Max Activating Token Index: 708

Text #13

Max Range: 3.1494. Min Range: -3.1494

Max Act: 1.8937. Min Act: -0.0000

Data Index: 581726 (C4 (Web Text))

Max Activating Token Index: 751

Text #14

Max Range: 3.1494. Min Range: -3.1494

Max Act: 1.7801. Min Act: -0.0000

Data Index: 1365145 (Python Code)

Max Activating Token Index: 506

Text #15

Max Range: 3.1494. Min Range: -3.1494

Max Act: 1.8086. Min Act: -0.0000

Data Index: 285599 (C4 (Web Text))

Max Activating Token Index: 760

Text #16

Max Range: 3.1494. Min Range: -3.1494

Max Act: 1.8786. Min Act: -0.0000

Data Index: 300144 (C4 (Web Text))

Max Activating Token Index: 770

Text #17

Max Range: 3.1494. Min Range: -3.1494

Max Act: 1.7677. Min Act: -0.0000

Data Index: 165368 (C4 (Web Text))

Max Activating Token Index: 502

Text #18

Max Range: 3.1494. Min Range: -3.1494

Max Act: 1.8461. Min Act: -0.0000

Data Index: 1116253 (C4 (Web Text))

Max Activating Token Index: 692

Text #19

Max Range: 3.1494. Min Range: -3.1494

Max Act: 1.8587. Min Act: -0.0000

Data Index: 1426132 (Python Code)

Max Activating Token Index: 326