Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 636 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.3594. Min Act: -0.0005

Data Index: 522018 (C4 (Web Text))

Max Activating Token Index: 942

Text #1

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.3225. Min Act: -0.0005

Data Index: 891411 (C4 (Web Text))

Max Activating Token Index: 888

Text #2

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.3240. Min Act: -0.0005

Data Index: 940597 (C4 (Web Text))

Max Activating Token Index: 284

Text #3

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.3133. Min Act: -0.0005

Data Index: 610395 (C4 (Web Text))

Max Activating Token Index: 849

Text #4

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2946. Min Act: -0.0005

Data Index: 304048 (C4 (Web Text))

Max Activating Token Index: 353

Text #5

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2942. Min Act: -0.0005

Data Index: 186532 (C4 (Web Text))

Max Activating Token Index: 880

Text #6

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2648. Min Act: -0.0005

Data Index: 1034647 (C4 (Web Text))

Max Activating Token Index: 825

Text #7

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2692. Min Act: -0.0005

Data Index: 492789 (C4 (Web Text))

Max Activating Token Index: 515

Text #8

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2964. Min Act: -0.0005

Data Index: 461496 (C4 (Web Text))

Max Activating Token Index: 879

Text #9

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2612. Min Act: -0.0005

Data Index: 4043 (C4 (Web Text))

Max Activating Token Index: 548

Text #10

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2686. Min Act: -0.0005

Data Index: 472117 (C4 (Web Text))

Max Activating Token Index: 611

Text #11

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2651. Min Act: -0.0005

Data Index: 342518 (C4 (Web Text))

Max Activating Token Index: 782

Text #12

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2299. Min Act: -0.0005

Data Index: 430187 (C4 (Web Text))

Max Activating Token Index: 816

Text #13

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2539. Min Act: -0.0005

Data Index: 340849 (C4 (Web Text))

Max Activating Token Index: 760

Text #14

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2635. Min Act: -0.0005

Data Index: 1054093 (C4 (Web Text))

Max Activating Token Index: 33

Text #15

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2628. Min Act: -0.0005

Data Index: 877681 (C4 (Web Text))

Max Activating Token Index: 977

Text #16

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2514. Min Act: -0.0005

Data Index: 1042549 (C4 (Web Text))

Max Activating Token Index: 710

Text #17

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2586. Min Act: -0.0005

Data Index: 1271228 (C4 (Web Text))

Max Activating Token Index: 261

Text #18

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2582. Min Act: -0.0005

Data Index: 3621 (C4 (Web Text))

Max Activating Token Index: 72

Text #19

Max Range: 0.3594. Min Range: -0.3594

Max Act: 0.2554. Min Act: -0.0005

Data Index: 683452 (C4 (Web Text))

Max Activating Token Index: 589