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

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

Neuron 128 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9620. Min Act: -0.0000

Data Index: 652837 (C4 (Web Text))

Max Activating Token Index: 796

Text #1

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9372. Min Act: -0.0000

Data Index: 1101519 (C4 (Web Text))

Max Activating Token Index: 585

Text #2

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9299. Min Act: -0.0000

Data Index: 1432957 (Python Code)

Max Activating Token Index: 721

Text #3

Max Range: 1.0289. Min Range: -1.0289

Max Act: 1.0289. Min Act: -0.0000

Data Index: 1021866 (C4 (Web Text))

Max Activating Token Index: 1019

Text #4

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9434. Min Act: -0.0000

Data Index: 1545757 (Python Code)

Max Activating Token Index: 564

Text #5

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.8683. Min Act: -0.0000

Data Index: 139899 (C4 (Web Text))

Max Activating Token Index: 282

Text #6

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9588. Min Act: -0.0000

Data Index: 1340809 (C4 (Web Text))

Max Activating Token Index: 1011

Text #7

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9009. Min Act: -0.0000

Data Index: 134983 (C4 (Web Text))

Max Activating Token Index: 35

Text #8

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9207. Min Act: -0.0000

Data Index: 570210 (C4 (Web Text))

Max Activating Token Index: 475

Text #9

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9190. Min Act: -0.0000

Data Index: 1387716 (Python Code)

Max Activating Token Index: 847

Text #10

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.8553. Min Act: -0.0000

Data Index: 518423 (C4 (Web Text))

Max Activating Token Index: 864

Text #11

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9272. Min Act: -0.0000

Data Index: 954051 (C4 (Web Text))

Max Activating Token Index: 433

Text #12

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.8720. Min Act: -0.0000

Data Index: 645054 (C4 (Web Text))

Max Activating Token Index: 109

Text #13

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9238. Min Act: -0.0000

Data Index: 519138 (C4 (Web Text))

Max Activating Token Index: 24

Text #14

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.8749. Min Act: -0.0000

Data Index: 1239923 (C4 (Web Text))

Max Activating Token Index: 290

Text #15

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.9017. Min Act: -0.0000

Data Index: 18351 (C4 (Web Text))

Max Activating Token Index: 42

Text #16

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.8960. Min Act: -0.0000

Data Index: 1432578 (Python Code)

Max Activating Token Index: 805

Text #17

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.8412. Min Act: -0.0000

Data Index: 60556 (C4 (Web Text))

Max Activating Token Index: 211

Text #18

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.8905. Min Act: -0.0000

Data Index: 1493884 (Python Code)

Max Activating Token Index: 138

Text #19

Max Range: 1.0289. Min Range: -1.0289

Max Act: 0.8613. Min Act: -0.0000

Data Index: 259141 (C4 (Web Text))

Max Activating Token Index: 21