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

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

Neuron 3945 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: 0.2216. Min Range: -0.2216

Max Act: 0.2216. Min Act: -0.0000

Data Index: 1412646 (Python Code)

Max Activating Token Index: 727

Text #1

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.2144. Min Act: -0.0000

Data Index: 1372156 (Python Code)

Max Activating Token Index: 78

Text #2

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.2126. Min Act: -0.0000

Data Index: 564960 (C4 (Web Text))

Max Activating Token Index: 303

Text #3

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.2142. Min Act: -0.0000

Data Index: 984867 (C4 (Web Text))

Max Activating Token Index: 268

Text #4

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.2051. Min Act: -0.0000

Data Index: 1492344 (Python Code)

Max Activating Token Index: 800

Text #5

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.2082. Min Act: -0.0000

Data Index: 344699 (C4 (Web Text))

Max Activating Token Index: 400

Text #6

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.2087. Min Act: -0.0000

Data Index: 1311974 (C4 (Web Text))

Max Activating Token Index: 324

Text #7

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.2079. Min Act: -0.0000

Data Index: 548128 (C4 (Web Text))

Max Activating Token Index: 117

Text #8

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.2018. Min Act: -0.0000

Data Index: 1223207 (C4 (Web Text))

Max Activating Token Index: 135

Text #9

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1927. Min Act: -0.0000

Data Index: 1219203 (C4 (Web Text))

Max Activating Token Index: 557

Text #10

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1868. Min Act: 0.0000

Data Index: 1558697 (Python Code)

Max Activating Token Index: 41

Text #11

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1918. Min Act: -0.0000

Data Index: 1039581 (C4 (Web Text))

Max Activating Token Index: 593

Text #12

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1858. Min Act: -0.0000

Data Index: 1584369 (Python Code)

Max Activating Token Index: 353

Text #13

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1870. Min Act: -0.0000

Data Index: 1119910 (C4 (Web Text))

Max Activating Token Index: 737

Text #14

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1868. Min Act: -0.0000

Data Index: 562042 (C4 (Web Text))

Max Activating Token Index: 853

Text #15

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1800. Min Act: -0.0000

Data Index: 1625285 (Python Code)

Max Activating Token Index: 40

Text #16

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1784. Min Act: -0.0000

Data Index: 449023 (C4 (Web Text))

Max Activating Token Index: 902

Text #17

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1784. Min Act: -0.0000

Data Index: 1551493 (Python Code)

Max Activating Token Index: 66

Text #18

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1789. Min Act: -0.0000

Data Index: 1535758 (Python Code)

Max Activating Token Index: 657

Text #19

Max Range: 0.2216. Min Range: -0.2216

Max Act: 0.1712. Min Act: -0.0000

Data Index: 1379715 (Python Code)

Max Activating Token Index: 450