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

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

Neuron 2634 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: 0.8521. Min Range: -0.8521

Max Act: 0.8521. Min Act: -0.0001

Data Index: 259609 (C4 (Web Text))

Max Activating Token Index: 121

Text #1

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.7722. Min Act: -0.0000

Data Index: 473426 (C4 (Web Text))

Max Activating Token Index: 226

Text #2

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.8350. Min Act: -0.0000

Data Index: 218641 (C4 (Web Text))

Max Activating Token Index: 14

Text #3

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.8293. Min Act: -0.0000

Data Index: 1169380 (C4 (Web Text))

Max Activating Token Index: 133

Text #4

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.7492. Min Act: -0.0000

Data Index: 576514 (C4 (Web Text))

Max Activating Token Index: 1020

Text #5

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.7740. Min Act: -0.0000

Data Index: 61747 (C4 (Web Text))

Max Activating Token Index: 380

Text #6

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.7061. Min Act: -0.0000

Data Index: 1608126 (Python Code)

Max Activating Token Index: 14

Text #7

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.6861. Min Act: -0.0000

Data Index: 1092614 (C4 (Web Text))

Max Activating Token Index: 632

Text #8

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.7084. Min Act: -0.0000

Data Index: 463341 (C4 (Web Text))

Max Activating Token Index: 232

Text #9

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.6595. Min Act: -0.0000

Data Index: 143750 (C4 (Web Text))

Max Activating Token Index: 182

Text #10

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.6065. Min Act: -0.0000

Data Index: 1152283 (C4 (Web Text))

Max Activating Token Index: 772

Text #11

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.7247. Min Act: -0.0000

Data Index: 1363004 (Python Code)

Max Activating Token Index: 26

Text #12

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.7165. Min Act: -0.0000

Data Index: 198103 (C4 (Web Text))

Max Activating Token Index: 457

Text #13

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.6984. Min Act: -0.0001

Data Index: 116091 (C4 (Web Text))

Max Activating Token Index: 77

Text #14

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.6465. Min Act: -0.0000

Data Index: 19590 (C4 (Web Text))

Max Activating Token Index: 97

Text #15

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.7076. Min Act: -0.0000

Data Index: 1003955 (C4 (Web Text))

Max Activating Token Index: 730

Text #16

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.6011. Min Act: -0.0000

Data Index: 269706 (C4 (Web Text))

Max Activating Token Index: 931

Text #17

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.6838. Min Act: -0.0000

Data Index: 22556 (C4 (Web Text))

Max Activating Token Index: 338

Text #18

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.6538. Min Act: -0.0000

Data Index: 1088965 (C4 (Web Text))

Max Activating Token Index: 28

Text #19

Max Range: 0.8521. Min Range: -0.8521

Max Act: 0.6664. Min Act: -0.0000

Data Index: 433591 (C4 (Web Text))

Max Activating Token Index: 565