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

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

Neuron 4750 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.8032. Min Act: -0.0000

Data Index: 6640 (C4 (Web Text))

Max Activating Token Index: 91

Text #1

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.6940. Min Act: -0.0000

Data Index: 1318966 (C4 (Web Text))

Max Activating Token Index: 129

Text #2

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.6773. Min Act: -0.0000

Data Index: 912508 (C4 (Web Text))

Max Activating Token Index: 916

Text #3

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.6479. Min Act: -0.0000

Data Index: 1344609 (C4 (Web Text))

Max Activating Token Index: 443

Text #4

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.6199. Min Act: -0.0000

Data Index: 862315 (C4 (Web Text))

Max Activating Token Index: 29

Text #5

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.5066. Min Act: -0.0000

Data Index: 1094633 (C4 (Web Text))

Max Activating Token Index: 851

Text #6

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.5074. Min Act: -0.0000

Data Index: 1122573 (C4 (Web Text))

Max Activating Token Index: 246

Text #7

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.5703. Min Act: -0.0000

Data Index: 818765 (C4 (Web Text))

Max Activating Token Index: 608

Text #8

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.5311. Min Act: -0.0000

Data Index: 810970 (C4 (Web Text))

Max Activating Token Index: 159

Text #9

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.4326. Min Act: -0.0000

Data Index: 872553 (C4 (Web Text))

Max Activating Token Index: 748

Text #10

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.5246. Min Act: -0.0000

Data Index: 101966 (C4 (Web Text))

Max Activating Token Index: 368

Text #11

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.4669. Min Act: -0.0000

Data Index: 614567 (C4 (Web Text))

Max Activating Token Index: 222

Text #12

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.4984. Min Act: -0.0000

Data Index: 326947 (C4 (Web Text))

Max Activating Token Index: 170

Text #13

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.5055. Min Act: -0.0000

Data Index: 1205371 (C4 (Web Text))

Max Activating Token Index: 1013

Text #14

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.4830. Min Act: -0.0000

Data Index: 1042324 (C4 (Web Text))

Max Activating Token Index: 470

Text #15

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.4613. Min Act: -0.0000

Data Index: 58450 (C4 (Web Text))

Max Activating Token Index: 483

Text #16

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.4903. Min Act: -0.0000

Data Index: 812125 (C4 (Web Text))

Max Activating Token Index: 77

Text #17

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.4685. Min Act: -0.0000

Data Index: 1359071 (C4 (Web Text))

Max Activating Token Index: 785

Text #18

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.4916. Min Act: -0.0000

Data Index: 1149759 (C4 (Web Text))

Max Activating Token Index: 276

Text #19

Max Range: 0.8032. Min Range: -0.8032

Max Act: 0.4545. Min Act: -0.0000

Data Index: 560457 (C4 (Web Text))

Max Activating Token Index: 694