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

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

Neuron 921 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.8333. Min Act: -0.0000

Data Index: 1557045 (Python Code)

Max Activating Token Index: 739

Text #1

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.6512. Min Act: -0.0000

Data Index: 158706 (C4 (Web Text))

Max Activating Token Index: 512

Text #2

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5877. Min Act: -0.0000

Data Index: 997519 (C4 (Web Text))

Max Activating Token Index: 594

Text #3

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.6053. Min Act: -0.0000

Data Index: 26102 (C4 (Web Text))

Max Activating Token Index: 784

Text #4

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5770. Min Act: -0.0000

Data Index: 1478986 (Python Code)

Max Activating Token Index: 94

Text #5

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5390. Min Act: -0.0000

Data Index: 94709 (C4 (Web Text))

Max Activating Token Index: 290

Text #6

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5496. Min Act: -0.0000

Data Index: 1049218 (C4 (Web Text))

Max Activating Token Index: 934

Text #7

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5856. Min Act: -0.0000

Data Index: 1072115 (C4 (Web Text))

Max Activating Token Index: 408

Text #8

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5681. Min Act: -0.0000

Data Index: 842901 (C4 (Web Text))

Max Activating Token Index: 170

Text #9

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5441. Min Act: -0.0000

Data Index: 1001290 (C4 (Web Text))

Max Activating Token Index: 567

Text #10

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.6020. Min Act: -0.0000

Data Index: 726938 (C4 (Web Text))

Max Activating Token Index: 237

Text #11

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5794. Min Act: -0.0000

Data Index: 887698 (C4 (Web Text))

Max Activating Token Index: 174

Text #12

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5483. Min Act: -0.0000

Data Index: 1268743 (C4 (Web Text))

Max Activating Token Index: 405

Text #13

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5252. Min Act: -0.0000

Data Index: 927425 (C4 (Web Text))

Max Activating Token Index: 710

Text #14

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.4887. Min Act: -0.0000

Data Index: 183520 (C4 (Web Text))

Max Activating Token Index: 935

Text #15

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5026. Min Act: -0.0000

Data Index: 1274334 (C4 (Web Text))

Max Activating Token Index: 1002

Text #16

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5673. Min Act: -0.0000

Data Index: 1192028 (C4 (Web Text))

Max Activating Token Index: 305

Text #17

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5363. Min Act: -0.0000

Data Index: 606006 (C4 (Web Text))

Max Activating Token Index: 806

Text #18

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5124. Min Act: -0.0000

Data Index: 1288739 (C4 (Web Text))

Max Activating Token Index: 709

Text #19

Max Range: 0.8333. Min Range: -0.8333

Max Act: 0.5444. Min Act: -0.0000

Data Index: 1411443 (Python Code)

Max Activating Token Index: 877