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

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

Neuron 4264 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: 1.4726. Min Range: -1.4726

Max Act: 1.3907. Min Act: -0.0000

Data Index: 573808 (C4 (Web Text))

Max Activating Token Index: 152

Text #1

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.4726. Min Act: -0.0000

Data Index: 1122524 (C4 (Web Text))

Max Activating Token Index: 797

Text #2

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.2156. Min Act: -0.0000

Data Index: 1490386 (Python Code)

Max Activating Token Index: 907

Text #3

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.3909. Min Act: -0.0000

Data Index: 976229 (C4 (Web Text))

Max Activating Token Index: 802

Text #4

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.3984. Min Act: -0.0000

Data Index: 1528663 (Python Code)

Max Activating Token Index: 275

Text #5

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.3123. Min Act: -0.0000

Data Index: 541839 (C4 (Web Text))

Max Activating Token Index: 177

Text #6

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.3096. Min Act: -0.0001

Data Index: 1432475 (Python Code)

Max Activating Token Index: 1014

Text #7

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.2939. Min Act: -0.0000

Data Index: 978779 (C4 (Web Text))

Max Activating Token Index: 59

Text #8

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.2288. Min Act: -0.0001

Data Index: 297508 (C4 (Web Text))

Max Activating Token Index: 837

Text #9

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.2283. Min Act: -0.0000

Data Index: 1503029 (Python Code)

Max Activating Token Index: 457

Text #10

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.2374. Min Act: -0.0000

Data Index: 125955 (C4 (Web Text))

Max Activating Token Index: 860

Text #11

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.2770. Min Act: -0.0000

Data Index: 1649132 (Python Code)

Max Activating Token Index: 683

Text #12

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.1884. Min Act: -0.0001

Data Index: 166662 (C4 (Web Text))

Max Activating Token Index: 510

Text #13

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.0569. Min Act: -0.0000

Data Index: 1402385 (Python Code)

Max Activating Token Index: 605

Text #14

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.1008. Min Act: -0.0000

Data Index: 1584433 (Python Code)

Max Activating Token Index: 322

Text #15

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.2058. Min Act: -0.0000

Data Index: 763183 (C4 (Web Text))

Max Activating Token Index: 455

Text #16

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.1604. Min Act: -0.0000

Data Index: 1313525 (C4 (Web Text))

Max Activating Token Index: 516

Text #17

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.0408. Min Act: -0.0001

Data Index: 151194 (C4 (Web Text))

Max Activating Token Index: 935

Text #18

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.0643. Min Act: -0.0001

Data Index: 1538606 (Python Code)

Max Activating Token Index: 537

Text #19

Max Range: 1.4726. Min Range: -1.4726

Max Act: 1.0973. Min Act: -0.0001

Data Index: 1283418 (C4 (Web Text))

Max Activating Token Index: 119