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

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

Neuron 3463 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5963. Min Act: -0.0001

Data Index: 115746 (C4 (Web Text))

Max Activating Token Index: 21

Text #1

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5755. Min Act: -0.0001

Data Index: 134984 (C4 (Web Text))

Max Activating Token Index: 867

Text #2

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5584. Min Act: -0.0001

Data Index: 498882 (C4 (Web Text))

Max Activating Token Index: 124

Text #3

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5639. Min Act: -0.0001

Data Index: 14038 (C4 (Web Text))

Max Activating Token Index: 750

Text #4

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5267. Min Act: -0.0001

Data Index: 899921 (C4 (Web Text))

Max Activating Token Index: 343

Text #5

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5206. Min Act: -0.0001

Data Index: 216621 (C4 (Web Text))

Max Activating Token Index: 39

Text #6

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5029. Min Act: -0.0001

Data Index: 304301 (C4 (Web Text))

Max Activating Token Index: 881

Text #7

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5248. Min Act: -0.0001

Data Index: 1218104 (C4 (Web Text))

Max Activating Token Index: 247

Text #8

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5184. Min Act: -0.0001

Data Index: 728639 (C4 (Web Text))

Max Activating Token Index: 145

Text #9

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5310. Min Act: -0.0001

Data Index: 1262850 (C4 (Web Text))

Max Activating Token Index: 572

Text #10

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5378. Min Act: -0.0001

Data Index: 297480 (C4 (Web Text))

Max Activating Token Index: 147

Text #11

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5067. Min Act: -0.0001

Data Index: 1007643 (C4 (Web Text))

Max Activating Token Index: 51

Text #12

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5739. Min Act: -0.0001

Data Index: 529557 (C4 (Web Text))

Max Activating Token Index: 273

Text #13

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.4843. Min Act: -0.0001

Data Index: 309064 (C4 (Web Text))

Max Activating Token Index: 29

Text #14

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5458. Min Act: -0.0001

Data Index: 982232 (C4 (Web Text))

Max Activating Token Index: 108

Text #15

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.4922. Min Act: -0.0001

Data Index: 808088 (C4 (Web Text))

Max Activating Token Index: 799

Text #16

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.5057. Min Act: -0.0001

Data Index: 297227 (C4 (Web Text))

Max Activating Token Index: 120

Text #17

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.4940. Min Act: -0.0001

Data Index: 1181358 (C4 (Web Text))

Max Activating Token Index: 869

Text #18

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.4799. Min Act: -0.0001

Data Index: 1115744 (C4 (Web Text))

Max Activating Token Index: 111

Text #19

Max Range: 0.5963. Min Range: -0.5963

Max Act: 0.4737. Min Act: -0.0001

Data Index: 721143 (C4 (Web Text))

Max Activating Token Index: 324