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

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

Neuron 4575 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.3219. Min Act: -0.0000

Data Index: 401667 (C4 (Web Text))

Max Activating Token Index: 77

Text #1

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.3679. Min Act: -0.0000

Data Index: 156985 (C4 (Web Text))

Max Activating Token Index: 789

Text #2

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.3161. Min Act: -0.0000

Data Index: 382515 (C4 (Web Text))

Max Activating Token Index: 915

Text #3

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.3003. Min Act: -0.0000

Data Index: 1070660 (C4 (Web Text))

Max Activating Token Index: 784

Text #4

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2969. Min Act: -0.0000

Data Index: 236422 (C4 (Web Text))

Max Activating Token Index: 810

Text #5

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.3073. Min Act: -0.0000

Data Index: 1226084 (C4 (Web Text))

Max Activating Token Index: 641

Text #6

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2747. Min Act: -0.0000

Data Index: 603844 (C4 (Web Text))

Max Activating Token Index: 727

Text #7

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2689. Min Act: -0.0000

Data Index: 268264 (C4 (Web Text))

Max Activating Token Index: 74

Text #8

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2745. Min Act: -0.0001

Data Index: 625023 (C4 (Web Text))

Max Activating Token Index: 682

Text #9

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2806. Min Act: -0.0000

Data Index: 450362 (C4 (Web Text))

Max Activating Token Index: 776

Text #10

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2723. Min Act: -0.0000

Data Index: 595393 (C4 (Web Text))

Max Activating Token Index: 939

Text #11

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2484. Min Act: -0.0001

Data Index: 1313938 (C4 (Web Text))

Max Activating Token Index: 171

Text #12

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2672. Min Act: -0.0000

Data Index: 1396772 (Python Code)

Max Activating Token Index: 471

Text #13

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2651. Min Act: -0.0000

Data Index: 624788 (C4 (Web Text))

Max Activating Token Index: 989

Text #14

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2481. Min Act: -0.0000

Data Index: 589011 (C4 (Web Text))

Max Activating Token Index: 686

Text #15

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2536. Min Act: -0.0000

Data Index: 725253 (C4 (Web Text))

Max Activating Token Index: 547

Text #16

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2550. Min Act: -0.0000

Data Index: 80388 (C4 (Web Text))

Max Activating Token Index: 16

Text #17

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2713. Min Act: -0.0000

Data Index: 1259953 (C4 (Web Text))

Max Activating Token Index: 301

Text #18

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2685. Min Act: -0.0000

Data Index: 1036609 (C4 (Web Text))

Max Activating Token Index: 1011

Text #19

Max Range: 0.3679. Min Range: -0.3679

Max Act: 0.2314. Min Act: -0.0001

Data Index: 961198 (C4 (Web Text))

Max Activating Token Index: 936