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

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

Neuron 2192 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: 0.8750. Min Range: -0.8750

Max Act: 0.8750. Min Act: -0.0000

Data Index: 1514637 (Python Code)

Max Activating Token Index: 307

Text #1

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.7632. Min Act: -0.0000

Data Index: 1476965 (Python Code)

Max Activating Token Index: 713

Text #2

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.7238. Min Act: -0.0000

Data Index: 685575 (C4 (Web Text))

Max Activating Token Index: 52

Text #3

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.6396. Min Act: -0.0000

Data Index: 1489067 (Python Code)

Max Activating Token Index: 204

Text #4

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.6287. Min Act: -0.0000

Data Index: 1197800 (C4 (Web Text))

Max Activating Token Index: 869

Text #5

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.6317. Min Act: -0.0000

Data Index: 730271 (C4 (Web Text))

Max Activating Token Index: 775

Text #6

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.6080. Min Act: -0.0000

Data Index: 1084010 (C4 (Web Text))

Max Activating Token Index: 139

Text #7

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.6595. Min Act: -0.0000

Data Index: 440247 (C4 (Web Text))

Max Activating Token Index: 226

Text #8

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.6260. Min Act: -0.0000

Data Index: 1187270 (C4 (Web Text))

Max Activating Token Index: 26

Text #9

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.6013. Min Act: -0.0000

Data Index: 118752 (C4 (Web Text))

Max Activating Token Index: 129

Text #10

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.5745. Min Act: -0.0000

Data Index: 1457999 (Python Code)

Max Activating Token Index: 233

Text #11

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.6001. Min Act: -0.0000

Data Index: 110364 (C4 (Web Text))

Max Activating Token Index: 109

Text #12

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.5931. Min Act: -0.0000

Data Index: 337915 (C4 (Web Text))

Max Activating Token Index: 399

Text #13

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.5977. Min Act: -0.0000

Data Index: 569839 (C4 (Web Text))

Max Activating Token Index: 730

Text #14

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.5872. Min Act: -0.0000

Data Index: 117020 (C4 (Web Text))

Max Activating Token Index: 29

Text #15

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.5446. Min Act: -0.0000

Data Index: 1656122 (Python Code)

Max Activating Token Index: 346

Text #16

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.5663. Min Act: -0.0000

Data Index: 803446 (C4 (Web Text))

Max Activating Token Index: 916

Text #17

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.5927. Min Act: -0.0000

Data Index: 1467465 (Python Code)

Max Activating Token Index: 361

Text #18

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.5710. Min Act: -0.0000

Data Index: 878714 (C4 (Web Text))

Max Activating Token Index: 889

Text #19

Max Range: 0.8750. Min Range: -0.8750

Max Act: 0.5439. Min Act: -0.0000

Data Index: 1636743 (Python Code)

Max Activating Token Index: 277