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

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

Neuron 3400 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.8090. Min Act: -0.0000

Data Index: 754008 (C4 (Web Text))

Max Activating Token Index: 443

Text #1

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.5948. Min Act: -0.0000

Data Index: 1254696 (C4 (Web Text))

Max Activating Token Index: 721

Text #2

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.6108. Min Act: 0.0002

Data Index: 1533629 (Python Code)

Max Activating Token Index: 107

Text #3

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.7111. Min Act: -0.0000

Data Index: 486830 (C4 (Web Text))

Max Activating Token Index: 834

Text #4

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.7030. Min Act: -0.0000

Data Index: 540629 (C4 (Web Text))

Max Activating Token Index: 393

Text #5

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.6469. Min Act: -0.0000

Data Index: 1457720 (Python Code)

Max Activating Token Index: 770

Text #6

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.6788. Min Act: -0.0000

Data Index: 1395556 (Python Code)

Max Activating Token Index: 94

Text #7

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.5910. Min Act: 0.0000

Data Index: 1578573 (Python Code)

Max Activating Token Index: 34

Text #8

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.6199. Min Act: -0.0000

Data Index: 543210 (C4 (Web Text))

Max Activating Token Index: 470

Text #9

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.5481. Min Act: -0.0000

Data Index: 1025232 (C4 (Web Text))

Max Activating Token Index: 733

Text #10

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.5933. Min Act: 0.0002

Data Index: 1380599 (Python Code)

Max Activating Token Index: 226

Text #11

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.5809. Min Act: -0.0000

Data Index: 1566098 (Python Code)

Max Activating Token Index: 416

Text #12

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.5629. Min Act: 0.0001

Data Index: 1491886 (Python Code)

Max Activating Token Index: 412

Text #13

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.6018. Min Act: 0.0000

Data Index: 1552865 (Python Code)

Max Activating Token Index: 58

Text #14

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.4998. Min Act: 0.0002

Data Index: 1513425 (Python Code)

Max Activating Token Index: 449

Text #15

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.5315. Min Act: -0.0000

Data Index: 1516364 (Python Code)

Max Activating Token Index: 47

Text #16

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.4926. Min Act: -0.0000

Data Index: 1654777 (Python Code)

Max Activating Token Index: 120

Text #17

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.4457. Min Act: 0.0001

Data Index: 1485016 (Python Code)

Max Activating Token Index: 221

Text #18

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.5282. Min Act: -0.0000

Data Index: 1142715 (C4 (Web Text))

Max Activating Token Index: 700

Text #19

Max Range: 1.8090. Min Range: -1.8090

Max Act: 1.4463. Min Act: 0.0000

Data Index: 1509056 (Python Code)

Max Activating Token Index: 45