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

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

Neuron 725 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: 1.9217. Min Range: -1.9217

Max Act: 1.9217. Min Act: -0.0000

Data Index: 290629 (C4 (Web Text))

Max Activating Token Index: 10

Text #1

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.6089. Min Act: -0.0000

Data Index: 1079676 (C4 (Web Text))

Max Activating Token Index: 17

Text #2

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.4922. Min Act: -0.0000

Data Index: 968148 (C4 (Web Text))

Max Activating Token Index: 431

Text #3

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.5701. Min Act: -0.0000

Data Index: 262554 (C4 (Web Text))

Max Activating Token Index: 541

Text #4

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.6048. Min Act: -0.0000

Data Index: 590491 (C4 (Web Text))

Max Activating Token Index: 216

Text #5

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.2127. Min Act: -0.0000

Data Index: 972605 (C4 (Web Text))

Max Activating Token Index: 1013

Text #6

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.3867. Min Act: -0.0000

Data Index: 952685 (C4 (Web Text))

Max Activating Token Index: 983

Text #7

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.3819. Min Act: -0.0000

Data Index: 1055652 (C4 (Web Text))

Max Activating Token Index: 460

Text #8

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.1442. Min Act: -0.0000

Data Index: 4500 (C4 (Web Text))

Max Activating Token Index: 878

Text #9

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.2288. Min Act: -0.0000

Data Index: 1365703 (Python Code)

Max Activating Token Index: 659

Text #10

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.1735. Min Act: -0.0000

Data Index: 1200713 (C4 (Web Text))

Max Activating Token Index: 479

Text #11

Max Range: 1.9217. Min Range: -1.9217

Max Act: 0.9444. Min Act: -0.0000

Data Index: 1610014 (Python Code)

Max Activating Token Index: 902

Text #12

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.0566. Min Act: -0.0000

Data Index: 396869 (C4 (Web Text))

Max Activating Token Index: 15

Text #13

Max Range: 1.9217. Min Range: -1.9217

Max Act: 0.9991. Min Act: -0.0000

Data Index: 348450 (C4 (Web Text))

Max Activating Token Index: 886

Text #14

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.0973. Min Act: -0.0000

Data Index: 1458348 (Python Code)

Max Activating Token Index: 280

Text #15

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.1710. Min Act: -0.0000

Data Index: 70284 (C4 (Web Text))

Max Activating Token Index: 83

Text #16

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.1091. Min Act: -0.0000

Data Index: 464670 (C4 (Web Text))

Max Activating Token Index: 809

Text #17

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.2239. Min Act: -0.0000

Data Index: 886833 (C4 (Web Text))

Max Activating Token Index: 858

Text #18

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.1278. Min Act: -0.0000

Data Index: 415708 (C4 (Web Text))

Max Activating Token Index: 498

Text #19

Max Range: 1.9217. Min Range: -1.9217

Max Act: 1.0338. Min Act: -0.0000

Data Index: 1621302 (Python Code)

Max Activating Token Index: 906