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

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

Neuron 1620 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: 0.6959. Min Range: -0.6959

Max Act: 0.6295. Min Act: -0.0000

Data Index: 433236 (C4 (Web Text))

Max Activating Token Index: 57

Text #1

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.6510. Min Act: -0.0000

Data Index: 318872 (C4 (Web Text))

Max Activating Token Index: 725

Text #2

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.6959. Min Act: -0.0000

Data Index: 836619 (C4 (Web Text))

Max Activating Token Index: 903

Text #3

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.6712. Min Act: -0.0000

Data Index: 1341584 (C4 (Web Text))

Max Activating Token Index: 59

Text #4

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.6017. Min Act: -0.0000

Data Index: 539569 (C4 (Web Text))

Max Activating Token Index: 773

Text #5

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.6020. Min Act: -0.0000

Data Index: 783427 (C4 (Web Text))

Max Activating Token Index: 11

Text #6

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5982. Min Act: -0.0000

Data Index: 1224260 (C4 (Web Text))

Max Activating Token Index: 989

Text #7

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5530. Min Act: -0.0000

Data Index: 634863 (C4 (Web Text))

Max Activating Token Index: 791

Text #8

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.6220. Min Act: -0.0000

Data Index: 341450 (C4 (Web Text))

Max Activating Token Index: 700

Text #9

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.6248. Min Act: -0.0000

Data Index: 1105353 (C4 (Web Text))

Max Activating Token Index: 152

Text #10

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.6095. Min Act: -0.0001

Data Index: 492481 (C4 (Web Text))

Max Activating Token Index: 941

Text #11

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5460. Min Act: -0.0000

Data Index: 928651 (C4 (Web Text))

Max Activating Token Index: 575

Text #12

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5986. Min Act: -0.0000

Data Index: 912993 (C4 (Web Text))

Max Activating Token Index: 742

Text #13

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5273. Min Act: -0.0000

Data Index: 195723 (C4 (Web Text))

Max Activating Token Index: 905

Text #14

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.4945. Min Act: -0.0000

Data Index: 777494 (C4 (Web Text))

Max Activating Token Index: 56

Text #15

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5427. Min Act: -0.0000

Data Index: 22627 (C4 (Web Text))

Max Activating Token Index: 996

Text #16

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5775. Min Act: -0.0001

Data Index: 466880 (C4 (Web Text))

Max Activating Token Index: 849

Text #17

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5129. Min Act: -0.0000

Data Index: 959479 (C4 (Web Text))

Max Activating Token Index: 611

Text #18

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5000. Min Act: -0.0000

Data Index: 384595 (C4 (Web Text))

Max Activating Token Index: 970

Text #19

Max Range: 0.6959. Min Range: -0.6959

Max Act: 0.5811. Min Act: -0.0000

Data Index: 359189 (C4 (Web Text))

Max Activating Token Index: 776