Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1576 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0751. Min Act: -0.0003

Data Index: 761048 (C4 (Web Text))

Max Activating Token Index: 816

Text #1

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0690. Min Act: -0.0003

Data Index: 291756 (C4 (Web Text))

Max Activating Token Index: 858

Text #2

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0681. Min Act: -0.0003

Data Index: 1469735 (Python Code)

Max Activating Token Index: 957

Text #3

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0693. Min Act: -0.0003

Data Index: 1439147 (Python Code)

Max Activating Token Index: 313

Text #4

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0691. Min Act: -0.0004

Data Index: 1157495 (C4 (Web Text))

Max Activating Token Index: 281

Text #5

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0655. Min Act: -0.0003

Data Index: 430436 (C4 (Web Text))

Max Activating Token Index: 115

Text #6

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0670. Min Act: -0.0003

Data Index: 691296 (C4 (Web Text))

Max Activating Token Index: 152

Text #7

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0671. Min Act: -0.0003

Data Index: 1540107 (Python Code)

Max Activating Token Index: 628

Text #8

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0664. Min Act: -0.0003

Data Index: 324552 (C4 (Web Text))

Max Activating Token Index: 53

Text #9

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0678. Min Act: -0.0003

Data Index: 1590600 (Python Code)

Max Activating Token Index: 387

Text #10

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0644. Min Act: -0.0003

Data Index: 36526 (C4 (Web Text))

Max Activating Token Index: 1007

Text #11

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0671. Min Act: -0.0002

Data Index: 1444705 (Python Code)

Max Activating Token Index: 609

Text #12

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0667. Min Act: -0.0003

Data Index: 219819 (C4 (Web Text))

Max Activating Token Index: 636

Text #13

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0684. Min Act: -0.0003

Data Index: 574034 (C4 (Web Text))

Max Activating Token Index: 80

Text #14

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0667. Min Act: -0.0004

Data Index: 526980 (C4 (Web Text))

Max Activating Token Index: 640

Text #15

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0644. Min Act: -0.0003

Data Index: 1091803 (C4 (Web Text))

Max Activating Token Index: 149

Text #16

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0669. Min Act: -0.0003

Data Index: 1043981 (C4 (Web Text))

Max Activating Token Index: 977

Text #17

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0631. Min Act: -0.0003

Data Index: 1644103 (Python Code)

Max Activating Token Index: 498

Text #18

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0657. Min Act: -0.0003

Data Index: 1219515 (C4 (Web Text))

Max Activating Token Index: 985

Text #19

Max Range: 0.0751. Min Range: -0.0751

Max Act: 0.0651. Min Act: -0.0003

Data Index: 573922 (C4 (Web Text))

Max Activating Token Index: 997