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

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

Neuron 3309 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.9029. Min Act: -0.0000

Data Index: 365362 (C4 (Web Text))

Max Activating Token Index: 34

Text #1

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.8779. Min Act: -0.0000

Data Index: 111728 (C4 (Web Text))

Max Activating Token Index: 656

Text #2

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.8874. Min Act: -0.0000

Data Index: 817430 (C4 (Web Text))

Max Activating Token Index: 328

Text #3

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.8381. Min Act: -0.0000

Data Index: 604956 (C4 (Web Text))

Max Activating Token Index: 196

Text #4

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.7628. Min Act: -0.0000

Data Index: 591018 (C4 (Web Text))

Max Activating Token Index: 293

Text #5

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.8254. Min Act: -0.0000

Data Index: 786286 (C4 (Web Text))

Max Activating Token Index: 954

Text #6

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.8686. Min Act: -0.0000

Data Index: 463816 (C4 (Web Text))

Max Activating Token Index: 240

Text #7

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.8134. Min Act: -0.0000

Data Index: 729035 (C4 (Web Text))

Max Activating Token Index: 516

Text #8

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.8186. Min Act: -0.0001

Data Index: 788709 (C4 (Web Text))

Max Activating Token Index: 873

Text #9

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.7715. Min Act: -0.0000

Data Index: 702346 (C4 (Web Text))

Max Activating Token Index: 937

Text #10

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.7995. Min Act: -0.0001

Data Index: 1398692 (Python Code)

Max Activating Token Index: 896

Text #11

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.8207. Min Act: -0.0000

Data Index: 29987 (C4 (Web Text))

Max Activating Token Index: 284

Text #12

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.7873. Min Act: -0.0000

Data Index: 1050558 (C4 (Web Text))

Max Activating Token Index: 835

Text #13

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.7167. Min Act: -0.0000

Data Index: 904938 (C4 (Web Text))

Max Activating Token Index: 103

Text #14

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.7128. Min Act: -0.0001

Data Index: 1618893 (Python Code)

Max Activating Token Index: 863

Text #15

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.7212. Min Act: -0.0000

Data Index: 298294 (C4 (Web Text))

Max Activating Token Index: 324

Text #16

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.6977. Min Act: -0.0000

Data Index: 1396360 (Python Code)

Max Activating Token Index: 434

Text #17

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.7389. Min Act: -0.0000

Data Index: 817541 (C4 (Web Text))

Max Activating Token Index: 232

Text #18

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.6874. Min Act: -0.0000

Data Index: 909716 (C4 (Web Text))

Max Activating Token Index: 995

Text #19

Max Range: 0.9029. Min Range: -0.9029

Max Act: 0.7073. Min Act: -0.0000

Data Index: 947426 (C4 (Web Text))

Max Activating Token Index: 809