Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3778 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('solu-12l')

Click the toggle to see the full text!

Text #0

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6854. Min Act: -0.0002

Data Index: 1522842 (Python Code)

Max Activating Token Index: 747

Text #1

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6743. Min Act: -0.0002

Data Index: 698870 (C4 (Web Text))

Max Activating Token Index: 167

Text #2

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6664. Min Act: -0.0002

Data Index: 195296 (C4 (Web Text))

Max Activating Token Index: 225

Text #3

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6825. Min Act: -0.0002

Data Index: 1613133 (Python Code)

Max Activating Token Index: 412

Text #4

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6754. Min Act: -0.0002

Data Index: 340643 (C4 (Web Text))

Max Activating Token Index: 154

Text #5

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6242. Min Act: -0.0002

Data Index: 764128 (C4 (Web Text))

Max Activating Token Index: 273

Text #6

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6300. Min Act: -0.0002

Data Index: 1381118 (Python Code)

Max Activating Token Index: 53

Text #7

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6309. Min Act: -0.0002

Data Index: 630097 (C4 (Web Text))

Max Activating Token Index: 805

Text #8

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6220. Min Act: -0.0002

Data Index: 1216956 (C4 (Web Text))

Max Activating Token Index: 132

Text #9

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.6063. Min Act: -0.0002

Data Index: 432657 (C4 (Web Text))

Max Activating Token Index: 107

Text #10

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5937. Min Act: -0.0002

Data Index: 1517780 (Python Code)

Max Activating Token Index: 624

Text #11

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5876. Min Act: -0.0002

Data Index: 920867 (C4 (Web Text))

Max Activating Token Index: 282

Text #12

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5892. Min Act: -0.0002

Data Index: 1650573 (Python Code)

Max Activating Token Index: 846

Text #13

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5848. Min Act: -0.0002

Data Index: 770011 (C4 (Web Text))

Max Activating Token Index: 386

Text #14

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5795. Min Act: -0.0002

Data Index: 583585 (C4 (Web Text))

Max Activating Token Index: 917

Text #15

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5655. Min Act: -0.0002

Data Index: 1587470 (Python Code)

Max Activating Token Index: 950

Text #16

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5727. Min Act: -0.0002

Data Index: 1042959 (C4 (Web Text))

Max Activating Token Index: 505

Text #17

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5695. Min Act: -0.0002

Data Index: 242719 (C4 (Web Text))

Max Activating Token Index: 767

Text #18

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5706. Min Act: -0.0002

Data Index: 1357378 (C4 (Web Text))

Max Activating Token Index: 589

Text #19

Max Range: 0.6854. Min Range: -0.6854

Max Act: 0.5602. Min Act: -0.0002

Data Index: 784854 (C4 (Web Text))

Max Activating Token Index: 961