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

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

Neuron 1500 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.2209. Min Act: -0.0001

Data Index: 402605 (C4 (Web Text))

Max Activating Token Index: 97

Text #1

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.2046. Min Act: -0.0001

Data Index: 511117 (C4 (Web Text))

Max Activating Token Index: 212

Text #2

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.2042. Min Act: -0.0001

Data Index: 960348 (C4 (Web Text))

Max Activating Token Index: 763

Text #3

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1891. Min Act: -0.0001

Data Index: 870620 (C4 (Web Text))

Max Activating Token Index: 373

Text #4

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1815. Min Act: -0.0001

Data Index: 1167032 (C4 (Web Text))

Max Activating Token Index: 728

Text #5

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1587. Min Act: -0.0001

Data Index: 1057555 (C4 (Web Text))

Max Activating Token Index: 781

Text #6

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1556. Min Act: -0.0001

Data Index: 869336 (C4 (Web Text))

Max Activating Token Index: 64

Text #7

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1655. Min Act: -0.0001

Data Index: 1042069 (C4 (Web Text))

Max Activating Token Index: 533

Text #8

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1468. Min Act: -0.0001

Data Index: 1337353 (C4 (Web Text))

Max Activating Token Index: 74

Text #9

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1460. Min Act: -0.0001

Data Index: 496707 (C4 (Web Text))

Max Activating Token Index: 671

Text #10

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1456. Min Act: -0.0001

Data Index: 993395 (C4 (Web Text))

Max Activating Token Index: 4

Text #11

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1394. Min Act: -0.0001

Data Index: 1036330 (C4 (Web Text))

Max Activating Token Index: 900

Text #12

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1439. Min Act: -0.0001

Data Index: 212942 (C4 (Web Text))

Max Activating Token Index: 486

Text #13

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1433. Min Act: -0.0001

Data Index: 792655 (C4 (Web Text))

Max Activating Token Index: 913

Text #14

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1421. Min Act: -0.0001

Data Index: 498466 (C4 (Web Text))

Max Activating Token Index: 861

Text #15

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1494. Min Act: -0.0001

Data Index: 555217 (C4 (Web Text))

Max Activating Token Index: 99

Text #16

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1470. Min Act: -0.0001

Data Index: 178279 (C4 (Web Text))

Max Activating Token Index: 130

Text #17

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1466. Min Act: -0.0001

Data Index: 643249 (C4 (Web Text))

Max Activating Token Index: 430

Text #18

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1422. Min Act: -0.0001

Data Index: 299954 (C4 (Web Text))

Max Activating Token Index: 477

Text #19

Max Range: 0.2209. Min Range: -0.2209

Max Act: 0.1422. Min Act: -0.0001

Data Index: 1465392 (Python Code)

Max Activating Token Index: 25