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

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

Neuron 3277 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: 0.7850. Min Range: -0.7850

Max Act: 0.7850. Min Act: -0.0000

Data Index: 941467 (C4 (Web Text))

Max Activating Token Index: 228

Text #1

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.7817. Min Act: -0.0000

Data Index: 835841 (C4 (Web Text))

Max Activating Token Index: 423

Text #2

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.4981. Min Act: -0.0000

Data Index: 452662 (C4 (Web Text))

Max Activating Token Index: 66

Text #3

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.6249. Min Act: -0.0000

Data Index: 678735 (C4 (Web Text))

Max Activating Token Index: 67

Text #4

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.6300. Min Act: -0.0000

Data Index: 1360537 (Python Code)

Max Activating Token Index: 227

Text #5

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.5089. Min Act: -0.0000

Data Index: 317413 (C4 (Web Text))

Max Activating Token Index: 788

Text #6

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.5258. Min Act: -0.0000

Data Index: 1476705 (Python Code)

Max Activating Token Index: 187

Text #7

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.4472. Min Act: -0.0000

Data Index: 1522232 (Python Code)

Max Activating Token Index: 300

Text #8

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.5178. Min Act: -0.0000

Data Index: 1413501 (Python Code)

Max Activating Token Index: 416

Text #9

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.3932. Min Act: -0.0000

Data Index: 668767 (C4 (Web Text))

Max Activating Token Index: 971

Text #10

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.4742. Min Act: -0.0000

Data Index: 623401 (C4 (Web Text))

Max Activating Token Index: 534

Text #11

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.3626. Min Act: -0.0000

Data Index: 1484598 (Python Code)

Max Activating Token Index: 753

Text #12

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.3959. Min Act: -0.0000

Data Index: 652277 (C4 (Web Text))

Max Activating Token Index: 677

Text #13

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.4500. Min Act: -0.0000

Data Index: 264959 (C4 (Web Text))

Max Activating Token Index: 743

Text #14

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.4152. Min Act: -0.0000

Data Index: 129294 (C4 (Web Text))

Max Activating Token Index: 515

Text #15

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.4607. Min Act: -0.0000

Data Index: 1562990 (Python Code)

Max Activating Token Index: 476

Text #16

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.3632. Min Act: -0.0000

Data Index: 1498848 (Python Code)

Max Activating Token Index: 778

Text #17

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.4051. Min Act: -0.0000

Data Index: 473014 (C4 (Web Text))

Max Activating Token Index: 795

Text #18

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.3867. Min Act: -0.0000

Data Index: 95409 (C4 (Web Text))

Max Activating Token Index: 859

Text #19

Max Range: 0.7850. Min Range: -0.7850

Max Act: 0.3626. Min Act: -0.0000

Data Index: 40182 (C4 (Web Text))

Max Activating Token Index: 618