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

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

Neuron 3950 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6854. Min Act: -0.0000

Data Index: 1464565 (Python Code)

Max Activating Token Index: 44

Text #1

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6084. Min Act: -0.0000

Data Index: 260407 (C4 (Web Text))

Max Activating Token Index: 150

Text #2

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5991. Min Act: -0.0000

Data Index: 1233008 (C4 (Web Text))

Max Activating Token Index: 142

Text #3

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6534. Min Act: -0.0000

Data Index: 476196 (C4 (Web Text))

Max Activating Token Index: 600

Text #4

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6184. Min Act: -0.0000

Data Index: 871004 (C4 (Web Text))

Max Activating Token Index: 971

Text #5

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6033. Min Act: -0.0000

Data Index: 1018224 (C4 (Web Text))

Max Activating Token Index: 405

Text #6

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6676. Min Act: -0.0000

Data Index: 9033 (C4 (Web Text))

Max Activating Token Index: 912

Text #7

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5600. Min Act: -0.0000

Data Index: 1162255 (C4 (Web Text))

Max Activating Token Index: 318

Text #8

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5593. Min Act: -0.0000

Data Index: 1566347 (Python Code)

Max Activating Token Index: 68

Text #9

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5768. Min Act: -0.0000

Data Index: 1393401 (Python Code)

Max Activating Token Index: 55

Text #10

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5694. Min Act: -0.0000

Data Index: 719619 (C4 (Web Text))

Max Activating Token Index: 380

Text #11

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5702. Min Act: -0.0000

Data Index: 673321 (C4 (Web Text))

Max Activating Token Index: 328

Text #12

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5323. Min Act: -0.0000

Data Index: 244947 (C4 (Web Text))

Max Activating Token Index: 478

Text #13

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5451. Min Act: -0.0000

Data Index: 144840 (C4 (Web Text))

Max Activating Token Index: 671

Text #14

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.4884. Min Act: -0.0000

Data Index: 578320 (C4 (Web Text))

Max Activating Token Index: 322

Text #15

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5545. Min Act: -0.0000

Data Index: 511452 (C4 (Web Text))

Max Activating Token Index: 608

Text #16

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5386. Min Act: -0.0000

Data Index: 402332 (C4 (Web Text))

Max Activating Token Index: 973

Text #17

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5480. Min Act: -0.0000

Data Index: 749988 (C4 (Web Text))

Max Activating Token Index: 375

Text #18

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5296. Min Act: -0.0000

Data Index: 647853 (C4 (Web Text))

Max Activating Token Index: 1019

Text #19

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.4856. Min Act: -0.0000

Data Index: 1136076 (C4 (Web Text))

Max Activating Token Index: 756