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

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

Neuron 3804 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.4934. Min Act: -0.0000

Data Index: 142982 (C4 (Web Text))

Max Activating Token Index: 616

Text #1

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.2459. Min Act: -0.0000

Data Index: 1135747 (C4 (Web Text))

Max Activating Token Index: 184

Text #2

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.1897. Min Act: -0.0000

Data Index: 1649249 (Python Code)

Max Activating Token Index: 413

Text #3

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.1949. Min Act: -0.0000

Data Index: 222658 (C4 (Web Text))

Max Activating Token Index: 727

Text #4

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.1379. Min Act: -0.0000

Data Index: 414039 (C4 (Web Text))

Max Activating Token Index: 692

Text #5

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.9884. Min Act: -0.0000

Data Index: 1115218 (C4 (Web Text))

Max Activating Token Index: 785

Text #6

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.0439. Min Act: -0.0000

Data Index: 35491 (C4 (Web Text))

Max Activating Token Index: 966

Text #7

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.0363. Min Act: -0.0000

Data Index: 864903 (C4 (Web Text))

Max Activating Token Index: 300

Text #8

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.0671. Min Act: -0.0000

Data Index: 1468473 (Python Code)

Max Activating Token Index: 138

Text #9

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.0612. Min Act: -0.0000

Data Index: 108245 (C4 (Web Text))

Max Activating Token Index: 372

Text #10

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.9435. Min Act: -0.0000

Data Index: 1219380 (C4 (Web Text))

Max Activating Token Index: 38

Text #11

Max Range: 1.4934. Min Range: -1.4934

Max Act: 1.0802. Min Act: -0.0000

Data Index: 1549018 (Python Code)

Max Activating Token Index: 961

Text #12

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.9920. Min Act: -0.0000

Data Index: 938591 (C4 (Web Text))

Max Activating Token Index: 115

Text #13

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.9121. Min Act: -0.0000

Data Index: 1446953 (Python Code)

Max Activating Token Index: 345

Text #14

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.8603. Min Act: -0.0000

Data Index: 442767 (C4 (Web Text))

Max Activating Token Index: 894

Text #15

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.8907. Min Act: -0.0000

Data Index: 314613 (C4 (Web Text))

Max Activating Token Index: 1018

Text #16

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.8907. Min Act: -0.0000

Data Index: 1654334 (Python Code)

Max Activating Token Index: 437

Text #17

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.8286. Min Act: -0.0000

Data Index: 752782 (C4 (Web Text))

Max Activating Token Index: 327

Text #18

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.8748. Min Act: -0.0000

Data Index: 1636850 (Python Code)

Max Activating Token Index: 572

Text #19

Max Range: 1.4934. Min Range: -1.4934

Max Act: 0.8721. Min Act: -0.0000

Data Index: 503168 (C4 (Web Text))

Max Activating Token Index: 246