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

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

Neuron 917 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.7334. Min Act: -0.0001

Data Index: 1599950 (Python Code)

Max Activating Token Index: 57

Text #1

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.4749. Min Act: -0.0001

Data Index: 287021 (C4 (Web Text))

Max Activating Token Index: 223

Text #2

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.4359. Min Act: -0.0001

Data Index: 1636796 (Python Code)

Max Activating Token Index: 251

Text #3

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.5101. Min Act: -0.0001

Data Index: 851597 (C4 (Web Text))

Max Activating Token Index: 707

Text #4

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.4665. Min Act: -0.0001

Data Index: 1416901 (Python Code)

Max Activating Token Index: 778

Text #5

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.5290. Min Act: -0.0001

Data Index: 208418 (C4 (Web Text))

Max Activating Token Index: 846

Text #6

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.2521. Min Act: -0.0001

Data Index: 1486190 (Python Code)

Max Activating Token Index: 895

Text #7

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.2981. Min Act: -0.0001

Data Index: 737090 (C4 (Web Text))

Max Activating Token Index: 409

Text #8

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.2077. Min Act: -0.0001

Data Index: 572969 (C4 (Web Text))

Max Activating Token Index: 952

Text #9

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.0841. Min Act: -0.0001

Data Index: 498029 (C4 (Web Text))

Max Activating Token Index: 776

Text #10

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.0802. Min Act: -0.0001

Data Index: 76820 (C4 (Web Text))

Max Activating Token Index: 911

Text #11

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.1803. Min Act: -0.0001

Data Index: 1521666 (Python Code)

Max Activating Token Index: 573

Text #12

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.2003. Min Act: -0.0001

Data Index: 1618703 (Python Code)

Max Activating Token Index: 810

Text #13

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.1600. Min Act: -0.0001

Data Index: 558673 (C4 (Web Text))

Max Activating Token Index: 5

Text #14

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.0825. Min Act: -0.0001

Data Index: 805304 (C4 (Web Text))

Max Activating Token Index: 656

Text #15

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.1710. Min Act: -0.0001

Data Index: 668413 (C4 (Web Text))

Max Activating Token Index: 610

Text #16

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.1221. Min Act: -0.0001

Data Index: 1159978 (C4 (Web Text))

Max Activating Token Index: 477

Text #17

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.1971. Min Act: -0.0001

Data Index: 1393544 (Python Code)

Max Activating Token Index: 741

Text #18

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.1950. Min Act: -0.0001

Data Index: 1648605 (Python Code)

Max Activating Token Index: 479

Text #19

Max Range: 1.7334. Min Range: -1.7334

Max Act: 1.0924. Min Act: -0.0001

Data Index: 310932 (C4 (Web Text))

Max Activating Token Index: 899