Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2145 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.8046. Min Act: -0.0001

Data Index: 491047 (C4 (Web Text))

Max Activating Token Index: 243

Text #1

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.6304. Min Act: -0.0002

Data Index: 316700 (C4 (Web Text))

Max Activating Token Index: 173

Text #2

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.7373. Min Act: -0.0001

Data Index: 5459 (C4 (Web Text))

Max Activating Token Index: 1021

Text #3

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.7078. Min Act: -0.0002

Data Index: 704833 (C4 (Web Text))

Max Activating Token Index: 969

Text #4

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.6680. Min Act: -0.0002

Data Index: 966056 (C4 (Web Text))

Max Activating Token Index: 900

Text #5

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.6585. Min Act: -0.0001

Data Index: 1643366 (Python Code)

Max Activating Token Index: 126

Text #6

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5796. Min Act: -0.0002

Data Index: 1326167 (C4 (Web Text))

Max Activating Token Index: 847

Text #7

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5832. Min Act: -0.0002

Data Index: 1227122 (C4 (Web Text))

Max Activating Token Index: 832

Text #8

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5697. Min Act: -0.0002

Data Index: 1011981 (C4 (Web Text))

Max Activating Token Index: 843

Text #9

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5595. Min Act: -0.0001

Data Index: 487272 (C4 (Web Text))

Max Activating Token Index: 524

Text #10

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5316. Min Act: -0.0002

Data Index: 1193251 (C4 (Web Text))

Max Activating Token Index: 831

Text #11

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5808. Min Act: -0.0001

Data Index: 1323474 (C4 (Web Text))

Max Activating Token Index: 857

Text #12

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.6097. Min Act: -0.0001

Data Index: 1459113 (Python Code)

Max Activating Token Index: 942

Text #13

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.6095. Min Act: -0.0002

Data Index: 872888 (C4 (Web Text))

Max Activating Token Index: 791

Text #14

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.6033. Min Act: -0.0002

Data Index: 1085801 (C4 (Web Text))

Max Activating Token Index: 262

Text #15

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5582. Min Act: -0.0001

Data Index: 475475 (C4 (Web Text))

Max Activating Token Index: 714

Text #16

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5844. Min Act: -0.0001

Data Index: 843768 (C4 (Web Text))

Max Activating Token Index: 668

Text #17

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5608. Min Act: -0.0002

Data Index: 827140 (C4 (Web Text))

Max Activating Token Index: 444

Text #18

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5058. Min Act: -0.0001

Data Index: 584873 (C4 (Web Text))

Max Activating Token Index: 968

Text #19

Max Range: 0.8046. Min Range: -0.8046

Max Act: 0.5536. Min Act: -0.0001

Data Index: 22885 (C4 (Web Text))

Max Activating Token Index: 478