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

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

Neuron 4768 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: 0.1080. Min Range: -0.1080

Max Act: 0.1080. Min Act: -0.0001

Data Index: 911319 (C4 (Web Text))

Max Activating Token Index: 432

Text #1

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.1037. Min Act: -0.0001

Data Index: 288741 (C4 (Web Text))

Max Activating Token Index: 307

Text #2

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0924. Min Act: -0.0001

Data Index: 54792 (C4 (Web Text))

Max Activating Token Index: 723

Text #3

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0951. Min Act: -0.0001

Data Index: 651231 (C4 (Web Text))

Max Activating Token Index: 278

Text #4

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0957. Min Act: -0.0001

Data Index: 332381 (C4 (Web Text))

Max Activating Token Index: 500

Text #5

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0954. Min Act: -0.0001

Data Index: 355749 (C4 (Web Text))

Max Activating Token Index: 987

Text #6

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0917. Min Act: -0.0001

Data Index: 397088 (C4 (Web Text))

Max Activating Token Index: 246

Text #7

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0919. Min Act: -0.0001

Data Index: 996949 (C4 (Web Text))

Max Activating Token Index: 261

Text #8

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0839. Min Act: -0.0001

Data Index: 386872 (C4 (Web Text))

Max Activating Token Index: 10

Text #9

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0932. Min Act: -0.0001

Data Index: 238255 (C4 (Web Text))

Max Activating Token Index: 147

Text #10

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0778. Min Act: -0.0001

Data Index: 812272 (C4 (Web Text))

Max Activating Token Index: 82

Text #11

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0828. Min Act: -0.0001

Data Index: 935327 (C4 (Web Text))

Max Activating Token Index: 559

Text #12

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0811. Min Act: -0.0001

Data Index: 984905 (C4 (Web Text))

Max Activating Token Index: 819

Text #13

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0855. Min Act: -0.0001

Data Index: 856450 (C4 (Web Text))

Max Activating Token Index: 89

Text #14

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0873. Min Act: -0.0001

Data Index: 589250 (C4 (Web Text))

Max Activating Token Index: 120

Text #15

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0816. Min Act: -0.0001

Data Index: 919148 (C4 (Web Text))

Max Activating Token Index: 329

Text #16

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0835. Min Act: -0.0001

Data Index: 1014302 (C4 (Web Text))

Max Activating Token Index: 553

Text #17

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0709. Min Act: -0.0001

Data Index: 1309697 (C4 (Web Text))

Max Activating Token Index: 804

Text #18

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0828. Min Act: -0.0001

Data Index: 602820 (C4 (Web Text))

Max Activating Token Index: 465

Text #19

Max Range: 0.1080. Min Range: -0.1080

Max Act: 0.0873. Min Act: -0.0001

Data Index: 1046852 (C4 (Web Text))

Max Activating Token Index: 283