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

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

Neuron 4492 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0562. Min Range: -1.0562

Max Act: 1.0467. Min Act: -0.0000

Data Index: 412852 (C4 (Web Text))

Max Activating Token Index: 523

Text #1

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.9547. Min Act: -0.0000

Data Index: 336344 (C4 (Web Text))

Max Activating Token Index: 955

Text #2

Max Range: 1.0562. Min Range: -1.0562

Max Act: 1.0351. Min Act: -0.0000

Data Index: 1098873 (C4 (Web Text))

Max Activating Token Index: 317

Text #3

Max Range: 1.0562. Min Range: -1.0562

Max Act: 1.0562. Min Act: -0.0000

Data Index: 398841 (C4 (Web Text))

Max Activating Token Index: 43

Text #4

Max Range: 1.0562. Min Range: -1.0562

Max Act: 1.0173. Min Act: -0.0000

Data Index: 64414 (C4 (Web Text))

Max Activating Token Index: 686

Text #5

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.9681. Min Act: -0.0000

Data Index: 791820 (C4 (Web Text))

Max Activating Token Index: 687

Text #6

Max Range: 1.0562. Min Range: -1.0562

Max Act: 1.0176. Min Act: -0.0000

Data Index: 625320 (C4 (Web Text))

Max Activating Token Index: 96

Text #7

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.9832. Min Act: -0.0000

Data Index: 91412 (C4 (Web Text))

Max Activating Token Index: 442

Text #8

Max Range: 1.0562. Min Range: -1.0562

Max Act: 1.0055. Min Act: -0.0000

Data Index: 386487 (C4 (Web Text))

Max Activating Token Index: 133

Text #9

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.9204. Min Act: -0.0000

Data Index: 810781 (C4 (Web Text))

Max Activating Token Index: 731

Text #10

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.9282. Min Act: -0.0000

Data Index: 987646 (C4 (Web Text))

Max Activating Token Index: 40

Text #11

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.8675. Min Act: -0.0000

Data Index: 744813 (C4 (Web Text))

Max Activating Token Index: 670

Text #12

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.8679. Min Act: -0.0000

Data Index: 1502873 (Python Code)

Max Activating Token Index: 1003

Text #13

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.8808. Min Act: -0.0000

Data Index: 676892 (C4 (Web Text))

Max Activating Token Index: 736

Text #14

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.8746. Min Act: -0.0000

Data Index: 509953 (C4 (Web Text))

Max Activating Token Index: 765

Text #15

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.8935. Min Act: -0.0000

Data Index: 824189 (C4 (Web Text))

Max Activating Token Index: 22

Text #16

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.8619. Min Act: -0.0000

Data Index: 1043083 (C4 (Web Text))

Max Activating Token Index: 172

Text #17

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.9105. Min Act: -0.0000

Data Index: 160113 (C4 (Web Text))

Max Activating Token Index: 313

Text #18

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.8620. Min Act: -0.0000

Data Index: 1254987 (C4 (Web Text))

Max Activating Token Index: 741

Text #19

Max Range: 1.0562. Min Range: -1.0562

Max Act: 0.9095. Min Act: -0.0000

Data Index: 988191 (C4 (Web Text))

Max Activating Token Index: 532