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

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

Neuron 4040 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.6627. Min Range: -2.6627

Max Act: 2.6627. Min Act: -0.0000

Data Index: 834076 (C4 (Web Text))

Max Activating Token Index: 618

Text #1

Max Range: 2.6627. Min Range: -2.6627

Max Act: 2.0788. Min Act: -0.0000

Data Index: 20107 (C4 (Web Text))

Max Activating Token Index: 878

Text #2

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.8720. Min Act: -0.0000

Data Index: 1345981 (C4 (Web Text))

Max Activating Token Index: 883

Text #3

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.8685. Min Act: -0.0000

Data Index: 1096691 (C4 (Web Text))

Max Activating Token Index: 424

Text #4

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.8078. Min Act: -0.0000

Data Index: 386710 (C4 (Web Text))

Max Activating Token Index: 553

Text #5

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.7738. Min Act: -0.0000

Data Index: 513594 (C4 (Web Text))

Max Activating Token Index: 670

Text #6

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.8490. Min Act: -0.0000

Data Index: 1449897 (Python Code)

Max Activating Token Index: 116

Text #7

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.8047. Min Act: -0.0000

Data Index: 1413276 (Python Code)

Max Activating Token Index: 869

Text #8

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.6382. Min Act: -0.0000

Data Index: 1644339 (Python Code)

Max Activating Token Index: 908

Text #9

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.6637. Min Act: -0.0000

Data Index: 1651388 (Python Code)

Max Activating Token Index: 962

Text #10

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.5547. Min Act: -0.0000

Data Index: 91384 (C4 (Web Text))

Max Activating Token Index: 368

Text #11

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.6792. Min Act: -0.0000

Data Index: 1561732 (Python Code)

Max Activating Token Index: 1000

Text #12

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.6514. Min Act: -0.0000

Data Index: 1221367 (C4 (Web Text))

Max Activating Token Index: 751

Text #13

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.8382. Min Act: -0.0000

Data Index: 1068882 (C4 (Web Text))

Max Activating Token Index: 314

Text #14

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.3064. Min Act: -0.0000

Data Index: 1461614 (Python Code)

Max Activating Token Index: 996

Text #15

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.5747. Min Act: -0.0000

Data Index: 603863 (C4 (Web Text))

Max Activating Token Index: 985

Text #16

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.5065. Min Act: -0.0000

Data Index: 1639914 (Python Code)

Max Activating Token Index: 915

Text #17

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.4289. Min Act: -0.0000

Data Index: 1454930 (Python Code)

Max Activating Token Index: 981

Text #18

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.5951. Min Act: -0.0000

Data Index: 1538929 (Python Code)

Max Activating Token Index: 962

Text #19

Max Range: 2.6627. Min Range: -2.6627

Max Act: 1.5588. Min Act: -0.0000

Data Index: 1586178 (Python Code)

Max Activating Token Index: 895