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

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

Neuron 1312 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0374. Min Act: -0.0002

Data Index: 959511 (C4 (Web Text))

Max Activating Token Index: 918

Text #1

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0312. Min Act: -0.0002

Data Index: 478779 (C4 (Web Text))

Max Activating Token Index: 310

Text #2

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0311. Min Act: -0.0002

Data Index: 736541 (C4 (Web Text))

Max Activating Token Index: 964

Text #3

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0279. Min Act: -0.0002

Data Index: 360267 (C4 (Web Text))

Max Activating Token Index: 278

Text #4

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0298. Min Act: -0.0002

Data Index: 944585 (C4 (Web Text))

Max Activating Token Index: 102

Text #5

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0267. Min Act: -0.0002

Data Index: 1269761 (C4 (Web Text))

Max Activating Token Index: 832

Text #6

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0305. Min Act: -0.0002

Data Index: 808180 (C4 (Web Text))

Max Activating Token Index: 965

Text #7

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0282. Min Act: -0.0002

Data Index: 1567034 (Python Code)

Max Activating Token Index: 657

Text #8

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0294. Min Act: -0.0002

Data Index: 408117 (C4 (Web Text))

Max Activating Token Index: 645

Text #9

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0301. Min Act: -0.0002

Data Index: 1033848 (C4 (Web Text))

Max Activating Token Index: 907

Text #10

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0301. Min Act: -0.0002

Data Index: 982049 (C4 (Web Text))

Max Activating Token Index: 816

Text #11

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0303. Min Act: -0.0002

Data Index: 1019162 (C4 (Web Text))

Max Activating Token Index: 835

Text #12

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0288. Min Act: -0.0002

Data Index: 941642 (C4 (Web Text))

Max Activating Token Index: 243

Text #13

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0287. Min Act: -0.0002

Data Index: 1113759 (C4 (Web Text))

Max Activating Token Index: 323

Text #14

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0303. Min Act: -0.0002

Data Index: 995741 (C4 (Web Text))

Max Activating Token Index: 822

Text #15

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0281. Min Act: -0.0002

Data Index: 1177573 (C4 (Web Text))

Max Activating Token Index: 64

Text #16

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0279. Min Act: -0.0002

Data Index: 344195 (C4 (Web Text))

Max Activating Token Index: 899

Text #17

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0332. Min Act: -0.0002

Data Index: 1076497 (C4 (Web Text))

Max Activating Token Index: 734

Text #18

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0281. Min Act: -0.0002

Data Index: 1563915 (Python Code)

Max Activating Token Index: 611

Text #19

Max Range: 0.0374. Min Range: -0.0374

Max Act: 0.0248. Min Act: -0.0002

Data Index: 314362 (C4 (Web Text))

Max Activating Token Index: 625