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

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

Neuron 1600 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: 1.1375. Min Range: -1.1375

Max Act: 1.1375. Min Act: -0.0000

Data Index: 1247087 (C4 (Web Text))

Max Activating Token Index: 370

Text #1

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.1238. Min Act: -0.0000

Data Index: 569486 (C4 (Web Text))

Max Activating Token Index: 672

Text #2

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.1369. Min Act: -0.0000

Data Index: 1508030 (Python Code)

Max Activating Token Index: 486

Text #3

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0963. Min Act: -0.0000

Data Index: 307051 (C4 (Web Text))

Max Activating Token Index: 386

Text #4

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0790. Min Act: -0.0000

Data Index: 611999 (C4 (Web Text))

Max Activating Token Index: 871

Text #5

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0898. Min Act: -0.0000

Data Index: 1157615 (C4 (Web Text))

Max Activating Token Index: 650

Text #6

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0592. Min Act: -0.0000

Data Index: 769973 (C4 (Web Text))

Max Activating Token Index: 94

Text #7

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0572. Min Act: -0.0000

Data Index: 361366 (C4 (Web Text))

Max Activating Token Index: 514

Text #8

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.1026. Min Act: -0.0000

Data Index: 1264467 (C4 (Web Text))

Max Activating Token Index: 135

Text #9

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0334. Min Act: -0.0000

Data Index: 922470 (C4 (Web Text))

Max Activating Token Index: 313

Text #10

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0354. Min Act: -0.0000

Data Index: 1494269 (Python Code)

Max Activating Token Index: 3

Text #11

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0944. Min Act: -0.0000

Data Index: 1106073 (C4 (Web Text))

Max Activating Token Index: 55

Text #12

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0551. Min Act: -0.0000

Data Index: 731392 (C4 (Web Text))

Max Activating Token Index: 656

Text #13

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0165. Min Act: -0.0000

Data Index: 45648 (C4 (Web Text))

Max Activating Token Index: 24

Text #14

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0453. Min Act: -0.0000

Data Index: 1132259 (C4 (Web Text))

Max Activating Token Index: 176

Text #15

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0580. Min Act: -0.0000

Data Index: 1276397 (C4 (Web Text))

Max Activating Token Index: 687

Text #16

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0549. Min Act: -0.0000

Data Index: 1410200 (Python Code)

Max Activating Token Index: 50

Text #17

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0873. Min Act: -0.0000

Data Index: 1072322 (C4 (Web Text))

Max Activating Token Index: 893

Text #18

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0006. Min Act: -0.0000

Data Index: 1526655 (Python Code)

Max Activating Token Index: 57

Text #19

Max Range: 1.1375. Min Range: -1.1375

Max Act: 1.0113. Min Act: -0.0000

Data Index: 316921 (C4 (Web Text))

Max Activating Token Index: 114