Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2552 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0914. Min Act: -0.0001

Data Index: 295729 (C4 (Web Text))

Max Activating Token Index: 526

Text #1

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0768. Min Act: -0.0001

Data Index: 874368 (C4 (Web Text))

Max Activating Token Index: 1007

Text #2

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0669. Min Act: -0.0001

Data Index: 987945 (C4 (Web Text))

Max Activating Token Index: 511

Text #3

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0726. Min Act: -0.0001

Data Index: 889971 (C4 (Web Text))

Max Activating Token Index: 328

Text #4

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0696. Min Act: -0.0001

Data Index: 807457 (C4 (Web Text))

Max Activating Token Index: 589

Text #5

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0702. Min Act: -0.0001

Data Index: 414451 (C4 (Web Text))

Max Activating Token Index: 469

Text #6

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0666. Min Act: -0.0001

Data Index: 1352139 (C4 (Web Text))

Max Activating Token Index: 534

Text #7

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0681. Min Act: -0.0001

Data Index: 640054 (C4 (Web Text))

Max Activating Token Index: 544

Text #8

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0673. Min Act: -0.0001

Data Index: 1549914 (Python Code)

Max Activating Token Index: 32

Text #9

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0621. Min Act: -0.0001

Data Index: 411187 (C4 (Web Text))

Max Activating Token Index: 51

Text #10

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0643. Min Act: -0.0001

Data Index: 316106 (C4 (Web Text))

Max Activating Token Index: 497

Text #11

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0659. Min Act: -0.0001

Data Index: 1561664 (Python Code)

Max Activating Token Index: 58

Text #12

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0606. Min Act: -0.0001

Data Index: 820538 (C4 (Web Text))

Max Activating Token Index: 433

Text #13

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0678. Min Act: -0.0001

Data Index: 862531 (C4 (Web Text))

Max Activating Token Index: 646

Text #14

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0633. Min Act: -0.0001

Data Index: 1521462 (Python Code)

Max Activating Token Index: 253

Text #15

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0644. Min Act: -0.0001

Data Index: 1545060 (Python Code)

Max Activating Token Index: 678

Text #16

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0627. Min Act: -0.0001

Data Index: 504436 (C4 (Web Text))

Max Activating Token Index: 301

Text #17

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0631. Min Act: -0.0001

Data Index: 788196 (C4 (Web Text))

Max Activating Token Index: 537

Text #18

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0679. Min Act: -0.0001

Data Index: 173010 (C4 (Web Text))

Max Activating Token Index: 734

Text #19

Max Range: 0.0914. Min Range: -0.0914

Max Act: 0.0631. Min Act: -0.0001

Data Index: 306854 (C4 (Web Text))

Max Activating Token Index: 596