Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4520 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.4860. Min Act: -0.0000

Data Index: 1411674 (Python Code)

Max Activating Token Index: 757

Text #1

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.4838. Min Act: -0.0000

Data Index: 1383083 (Python Code)

Max Activating Token Index: 528

Text #2

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.4595. Min Act: -0.0000

Data Index: 1142645 (C4 (Web Text))

Max Activating Token Index: 878

Text #3

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.4187. Min Act: -0.0000

Data Index: 155111 (C4 (Web Text))

Max Activating Token Index: 618

Text #4

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.4004. Min Act: -0.0000

Data Index: 1462751 (Python Code)

Max Activating Token Index: 562

Text #5

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3720. Min Act: -0.0000

Data Index: 1606674 (Python Code)

Max Activating Token Index: 680

Text #6

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.4045. Min Act: -0.0000

Data Index: 1418079 (Python Code)

Max Activating Token Index: 684

Text #7

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.4210. Min Act: -0.0000

Data Index: 226919 (C4 (Web Text))

Max Activating Token Index: 990

Text #8

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3895. Min Act: -0.0000

Data Index: 168157 (C4 (Web Text))

Max Activating Token Index: 691

Text #9

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3775. Min Act: -0.0000

Data Index: 1474117 (Python Code)

Max Activating Token Index: 470

Text #10

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.4022. Min Act: -0.0000

Data Index: 1333201 (C4 (Web Text))

Max Activating Token Index: 30

Text #11

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3569. Min Act: -0.0000

Data Index: 1461667 (Python Code)

Max Activating Token Index: 333

Text #12

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3467. Min Act: -0.0000

Data Index: 1620232 (Python Code)

Max Activating Token Index: 703

Text #13

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3561. Min Act: -0.0000

Data Index: 767678 (C4 (Web Text))

Max Activating Token Index: 149

Text #14

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3561. Min Act: -0.0000

Data Index: 1516093 (Python Code)

Max Activating Token Index: 966

Text #15

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3339. Min Act: -0.0000

Data Index: 1414998 (Python Code)

Max Activating Token Index: 678

Text #16

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3399. Min Act: -0.0000

Data Index: 1431688 (Python Code)

Max Activating Token Index: 1018

Text #17

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3418. Min Act: -0.0000

Data Index: 1642842 (Python Code)

Max Activating Token Index: 111

Text #18

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3101. Min Act: -0.0000

Data Index: 1625166 (Python Code)

Max Activating Token Index: 198

Text #19

Max Range: 0.4860. Min Range: -0.4860

Max Act: 0.3357. Min Act: -0.0000

Data Index: 1467572 (Python Code)

Max Activating Token Index: 56