Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3818 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.3465. Min Act: -0.0000

Data Index: 861878 (C4 (Web Text))

Max Activating Token Index: 682

Text #1

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2587. Min Act: -0.0000

Data Index: 1304845 (C4 (Web Text))

Max Activating Token Index: 34

Text #2

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2622. Min Act: -0.0000

Data Index: 885476 (C4 (Web Text))

Max Activating Token Index: 366

Text #3

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2109. Min Act: -0.0000

Data Index: 496134 (C4 (Web Text))

Max Activating Token Index: 795

Text #4

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2481. Min Act: -0.0000

Data Index: 1019783 (C4 (Web Text))

Max Activating Token Index: 922

Text #5

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2662. Min Act: -0.0000

Data Index: 662601 (C4 (Web Text))

Max Activating Token Index: 89

Text #6

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2346. Min Act: -0.0000

Data Index: 1436781 (Python Code)

Max Activating Token Index: 867

Text #7

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2359. Min Act: -0.0000

Data Index: 108902 (C4 (Web Text))

Max Activating Token Index: 16

Text #8

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2216. Min Act: -0.0000

Data Index: 257092 (C4 (Web Text))

Max Activating Token Index: 484

Text #9

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2097. Min Act: -0.0000

Data Index: 1297401 (C4 (Web Text))

Max Activating Token Index: 53

Text #10

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2053. Min Act: -0.0000

Data Index: 667536 (C4 (Web Text))

Max Activating Token Index: 26

Text #11

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2047. Min Act: -0.0000

Data Index: 905621 (C4 (Web Text))

Max Activating Token Index: 202

Text #12

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2246. Min Act: -0.0000

Data Index: 80403 (C4 (Web Text))

Max Activating Token Index: 564

Text #13

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2134. Min Act: -0.0000

Data Index: 226888 (C4 (Web Text))

Max Activating Token Index: 853

Text #14

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.1970. Min Act: -0.0000

Data Index: 499050 (C4 (Web Text))

Max Activating Token Index: 126

Text #15

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2066. Min Act: -0.0000

Data Index: 676539 (C4 (Web Text))

Max Activating Token Index: 282

Text #16

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2156. Min Act: -0.0000

Data Index: 754186 (C4 (Web Text))

Max Activating Token Index: 269

Text #17

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.1974. Min Act: -0.0000

Data Index: 806911 (C4 (Web Text))

Max Activating Token Index: 13

Text #18

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.2030. Min Act: -0.0000

Data Index: 1003435 (C4 (Web Text))

Max Activating Token Index: 947

Text #19

Max Range: 0.3465. Min Range: -0.3465

Max Act: 0.1909. Min Act: -0.0000

Data Index: 972130 (C4 (Web Text))

Max Activating Token Index: 928