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

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

Neuron 1266 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.6554. Min Act: -0.0000

Data Index: 719913 (C4 (Web Text))

Max Activating Token Index: 104

Text #1

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5398. Min Act: -0.0000

Data Index: 241963 (C4 (Web Text))

Max Activating Token Index: 948

Text #2

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.6005. Min Act: 0.0000

Data Index: 1277974 (C4 (Web Text))

Max Activating Token Index: 197

Text #3

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5785. Min Act: 0.0000

Data Index: 634288 (C4 (Web Text))

Max Activating Token Index: 484

Text #4

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5516. Min Act: -0.0000

Data Index: 460754 (C4 (Web Text))

Max Activating Token Index: 620

Text #5

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5326. Min Act: -0.0000

Data Index: 1469695 (Python Code)

Max Activating Token Index: 886

Text #6

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5125. Min Act: -0.0000

Data Index: 664862 (C4 (Web Text))

Max Activating Token Index: 541

Text #7

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5748. Min Act: 0.0000

Data Index: 590449 (C4 (Web Text))

Max Activating Token Index: 50

Text #8

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5442. Min Act: -0.0000

Data Index: 1641191 (Python Code)

Max Activating Token Index: 348

Text #9

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5489. Min Act: -0.0000

Data Index: 918713 (C4 (Web Text))

Max Activating Token Index: 370

Text #10

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5447. Min Act: -0.0000

Data Index: 588532 (C4 (Web Text))

Max Activating Token Index: 588

Text #11

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5827. Min Act: -0.0000

Data Index: 1448711 (Python Code)

Max Activating Token Index: 863

Text #12

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.6011. Min Act: -0.0000

Data Index: 1494333 (Python Code)

Max Activating Token Index: 197

Text #13

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5329. Min Act: -0.0000

Data Index: 1228444 (C4 (Web Text))

Max Activating Token Index: 727

Text #14

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5606. Min Act: -0.0000

Data Index: 1370711 (Python Code)

Max Activating Token Index: 950

Text #15

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5440. Min Act: -0.0000

Data Index: 1618672 (Python Code)

Max Activating Token Index: 134

Text #16

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.5116. Min Act: 0.0000

Data Index: 837450 (C4 (Web Text))

Max Activating Token Index: 341

Text #17

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.4647. Min Act: -0.0000

Data Index: 1603776 (Python Code)

Max Activating Token Index: 351

Text #18

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.4491. Min Act: -0.0000

Data Index: 1446775 (Python Code)

Max Activating Token Index: 822

Text #19

Max Range: 0.6554. Min Range: -0.6554

Max Act: 0.4742. Min Act: -0.0000

Data Index: 1032909 (C4 (Web Text))

Max Activating Token Index: 439