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

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

Neuron 595 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: 0.6966. Min Range: -0.6966

Max Act: 0.6936. Min Act: -0.0001

Data Index: 1075107 (C4 (Web Text))

Max Activating Token Index: 285

Text #1

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6966. Min Act: -0.0001

Data Index: 678844 (C4 (Web Text))

Max Activating Token Index: 718

Text #2

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6396. Min Act: -0.0001

Data Index: 478639 (C4 (Web Text))

Max Activating Token Index: 551

Text #3

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6710. Min Act: -0.0001

Data Index: 756047 (C4 (Web Text))

Max Activating Token Index: 1019

Text #4

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6267. Min Act: -0.0001

Data Index: 790125 (C4 (Web Text))

Max Activating Token Index: 74

Text #5

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6483. Min Act: -0.0002

Data Index: 61745 (C4 (Web Text))

Max Activating Token Index: 976

Text #6

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6256. Min Act: -0.0002

Data Index: 503745 (C4 (Web Text))

Max Activating Token Index: 794

Text #7

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6092. Min Act: -0.0001

Data Index: 944513 (C4 (Web Text))

Max Activating Token Index: 641

Text #8

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.5212. Min Act: -0.0002

Data Index: 1122258 (C4 (Web Text))

Max Activating Token Index: 923

Text #9

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6209. Min Act: -0.0002

Data Index: 955960 (C4 (Web Text))

Max Activating Token Index: 850

Text #10

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.5830. Min Act: -0.0002

Data Index: 612981 (C4 (Web Text))

Max Activating Token Index: 873

Text #11

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.5919. Min Act: -0.0002

Data Index: 1136135 (C4 (Web Text))

Max Activating Token Index: 996

Text #12

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6245. Min Act: -0.0001

Data Index: 599731 (C4 (Web Text))

Max Activating Token Index: 996

Text #13

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6031. Min Act: -0.0002

Data Index: 172484 (C4 (Web Text))

Max Activating Token Index: 150

Text #14

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.5714. Min Act: -0.0001

Data Index: 1279375 (C4 (Web Text))

Max Activating Token Index: 813

Text #15

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.6092. Min Act: -0.0002

Data Index: 1032620 (C4 (Web Text))

Max Activating Token Index: 199

Text #16

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.5945. Min Act: -0.0002

Data Index: 670998 (C4 (Web Text))

Max Activating Token Index: 548

Text #17

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.5676. Min Act: -0.0002

Data Index: 665341 (C4 (Web Text))

Max Activating Token Index: 400

Text #18

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.5419. Min Act: -0.0001

Data Index: 144728 (C4 (Web Text))

Max Activating Token Index: 938

Text #19

Max Range: 0.6966. Min Range: -0.6966

Max Act: 0.5797. Min Act: -0.0001

Data Index: 171592 (C4 (Web Text))

Max Activating Token Index: 462