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

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

Neuron 3663 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.8511. Min Act: -0.0000

Data Index: 288425 (C4 (Web Text))

Max Activating Token Index: 771

Text #1

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.6937. Min Act: -0.0000

Data Index: 74472 (C4 (Web Text))

Max Activating Token Index: 521

Text #2

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.6838. Min Act: -0.0000

Data Index: 97141 (C4 (Web Text))

Max Activating Token Index: 582

Text #3

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.6456. Min Act: -0.0000

Data Index: 350746 (C4 (Web Text))

Max Activating Token Index: 608

Text #4

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5951. Min Act: -0.0000

Data Index: 1029235 (C4 (Web Text))

Max Activating Token Index: 66

Text #5

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.6576. Min Act: -0.0000

Data Index: 600332 (C4 (Web Text))

Max Activating Token Index: 445

Text #6

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5884. Min Act: -0.0000

Data Index: 401266 (C4 (Web Text))

Max Activating Token Index: 740

Text #7

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.6043. Min Act: -0.0000

Data Index: 821556 (C4 (Web Text))

Max Activating Token Index: 775

Text #8

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5997. Min Act: -0.0000

Data Index: 59131 (C4 (Web Text))

Max Activating Token Index: 143

Text #9

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5579. Min Act: -0.0000

Data Index: 739548 (C4 (Web Text))

Max Activating Token Index: 728

Text #10

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.6445. Min Act: -0.0000

Data Index: 516862 (C4 (Web Text))

Max Activating Token Index: 781

Text #11

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5130. Min Act: -0.0000

Data Index: 416633 (C4 (Web Text))

Max Activating Token Index: 389

Text #12

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5509. Min Act: -0.0000

Data Index: 153710 (C4 (Web Text))

Max Activating Token Index: 566

Text #13

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5837. Min Act: -0.0000

Data Index: 208571 (C4 (Web Text))

Max Activating Token Index: 210

Text #14

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5228. Min Act: -0.0000

Data Index: 853952 (C4 (Web Text))

Max Activating Token Index: 314

Text #15

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5812. Min Act: -0.0000

Data Index: 317733 (C4 (Web Text))

Max Activating Token Index: 777

Text #16

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.4872. Min Act: -0.0000

Data Index: 337252 (C4 (Web Text))

Max Activating Token Index: 741

Text #17

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5354. Min Act: -0.0000

Data Index: 512992 (C4 (Web Text))

Max Activating Token Index: 998

Text #18

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5678. Min Act: -0.0000

Data Index: 1479839 (Python Code)

Max Activating Token Index: 870

Text #19

Max Range: 0.8511. Min Range: -0.8511

Max Act: 0.5003. Min Act: -0.0000

Data Index: 162744 (C4 (Web Text))

Max Activating Token Index: 90