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

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

Neuron 1658 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.3576. Min Act: -0.0000

Data Index: 924666 (C4 (Web Text))

Max Activating Token Index: 733

Text #1

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.2473. Min Act: -0.0000

Data Index: 536057 (C4 (Web Text))

Max Activating Token Index: 452

Text #2

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.2236. Min Act: -0.0000

Data Index: 945255 (C4 (Web Text))

Max Activating Token Index: 105

Text #3

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.1917. Min Act: -0.0000

Data Index: 858960 (C4 (Web Text))

Max Activating Token Index: 541

Text #4

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.1630. Min Act: -0.0000

Data Index: 945449 (C4 (Web Text))

Max Activating Token Index: 940

Text #5

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.1327. Min Act: -0.0000

Data Index: 1043087 (C4 (Web Text))

Max Activating Token Index: 982

Text #6

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.1742. Min Act: -0.0000

Data Index: 270198 (C4 (Web Text))

Max Activating Token Index: 526

Text #7

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.1643. Min Act: -0.0000

Data Index: 737820 (C4 (Web Text))

Max Activating Token Index: 773

Text #8

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.1195. Min Act: -0.0000

Data Index: 1207637 (C4 (Web Text))

Max Activating Token Index: 703

Text #9

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.1470. Min Act: -0.0000

Data Index: 1569469 (Python Code)

Max Activating Token Index: 758

Text #10

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0777. Min Act: -0.0000

Data Index: 111757 (C4 (Web Text))

Max Activating Token Index: 586

Text #11

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0955. Min Act: -0.0000

Data Index: 984369 (C4 (Web Text))

Max Activating Token Index: 44

Text #12

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0865. Min Act: -0.0000

Data Index: 462600 (C4 (Web Text))

Max Activating Token Index: 160

Text #13

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0782. Min Act: -0.0000

Data Index: 952527 (C4 (Web Text))

Max Activating Token Index: 611

Text #14

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0772. Min Act: -0.0000

Data Index: 1602601 (Python Code)

Max Activating Token Index: 679

Text #15

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0771. Min Act: -0.0000

Data Index: 620548 (C4 (Web Text))

Max Activating Token Index: 767

Text #16

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0380. Min Act: -0.0000

Data Index: 688538 (C4 (Web Text))

Max Activating Token Index: 660

Text #17

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0685. Min Act: -0.0000

Data Index: 42909 (C4 (Web Text))

Max Activating Token Index: 568

Text #18

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0119. Min Act: -0.0000

Data Index: 1435194 (Python Code)

Max Activating Token Index: 400

Text #19

Max Range: 1.3576. Min Range: -1.3576

Max Act: 1.0261. Min Act: -0.0000

Data Index: 175452 (C4 (Web Text))

Max Activating Token Index: 311