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

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

Neuron 1121 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.9742. Min Act: -0.0002

Data Index: 1599630 (Python Code)

Max Activating Token Index: 7

Text #1

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.9201. Min Act: -0.0002

Data Index: 1487367 (Python Code)

Max Activating Token Index: 7

Text #2

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.9201. Min Act: -0.0002

Data Index: 1390939 (Python Code)

Max Activating Token Index: 7

Text #3

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8520. Min Act: -0.0002

Data Index: 1385285 (Python Code)

Max Activating Token Index: 434

Text #4

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8426. Min Act: -0.0002

Data Index: 1644958 (Python Code)

Max Activating Token Index: 793

Text #5

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8563. Min Act: -0.0002

Data Index: 1611462 (Python Code)

Max Activating Token Index: 6

Text #6

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8722. Min Act: -0.0002

Data Index: 1633083 (Python Code)

Max Activating Token Index: 799

Text #7

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8487. Min Act: -0.0002

Data Index: 1528025 (Python Code)

Max Activating Token Index: 696

Text #8

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8391. Min Act: -0.0002

Data Index: 1503147 (Python Code)

Max Activating Token Index: 8

Text #9

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8094. Min Act: -0.0002

Data Index: 1493832 (Python Code)

Max Activating Token Index: 7

Text #10

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8042. Min Act: -0.0002

Data Index: 1545000 (Python Code)

Max Activating Token Index: 749

Text #11

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8418. Min Act: -0.0002

Data Index: 1599069 (Python Code)

Max Activating Token Index: 24

Text #12

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.7947. Min Act: -0.0002

Data Index: 1596151 (Python Code)

Max Activating Token Index: 86

Text #13

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.7839. Min Act: -0.0002

Data Index: 1583788 (Python Code)

Max Activating Token Index: 406

Text #14

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8622. Min Act: -0.0002

Data Index: 1414871 (Python Code)

Max Activating Token Index: 808

Text #15

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.7749. Min Act: -0.0002

Data Index: 1533257 (Python Code)

Max Activating Token Index: 22

Text #16

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.8084. Min Act: -0.0002

Data Index: 1476134 (Python Code)

Max Activating Token Index: 39

Text #17

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.7538. Min Act: -0.0002

Data Index: 1388515 (Python Code)

Max Activating Token Index: 659

Text #18

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.7736. Min Act: -0.0002

Data Index: 1364394 (Python Code)

Max Activating Token Index: 855

Text #19

Max Range: 0.9742. Min Range: -0.9742

Max Act: 0.7465. Min Act: -0.0002

Data Index: 1563580 (Python Code)

Max Activating Token Index: 357