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

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

Neuron 272 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.8053. Min Act: -0.0000

Data Index: 864145 (C4 (Web Text))

Max Activating Token Index: 636

Text #1

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7868. Min Act: -0.0000

Data Index: 1525518 (Python Code)

Max Activating Token Index: 221

Text #2

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.8102. Min Act: -0.0000

Data Index: 1321374 (C4 (Web Text))

Max Activating Token Index: 827

Text #3

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7399. Min Act: -0.0000

Data Index: 1202731 (C4 (Web Text))

Max Activating Token Index: 949

Text #4

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7652. Min Act: -0.0000

Data Index: 1625031 (Python Code)

Max Activating Token Index: 936

Text #5

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7599. Min Act: -0.0000

Data Index: 259440 (C4 (Web Text))

Max Activating Token Index: 150

Text #6

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7478. Min Act: -0.0000

Data Index: 1016704 (C4 (Web Text))

Max Activating Token Index: 532

Text #7

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7430. Min Act: -0.0000

Data Index: 599245 (C4 (Web Text))

Max Activating Token Index: 587

Text #8

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.6869. Min Act: -0.0000

Data Index: 792040 (C4 (Web Text))

Max Activating Token Index: 955

Text #9

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7817. Min Act: -0.0000

Data Index: 192367 (C4 (Web Text))

Max Activating Token Index: 709

Text #10

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7080. Min Act: -0.0000

Data Index: 1012388 (C4 (Web Text))

Max Activating Token Index: 460

Text #11

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7025. Min Act: -0.0000

Data Index: 968947 (C4 (Web Text))

Max Activating Token Index: 488

Text #12

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.7018. Min Act: -0.0000

Data Index: 1097397 (C4 (Web Text))

Max Activating Token Index: 511

Text #13

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.6891. Min Act: -0.0001

Data Index: 1364420 (Python Code)

Max Activating Token Index: 134

Text #14

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.6944. Min Act: -0.0000

Data Index: 46297 (C4 (Web Text))

Max Activating Token Index: 732

Text #15

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.6686. Min Act: -0.0000

Data Index: 1333397 (C4 (Web Text))

Max Activating Token Index: 1018

Text #16

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.6993. Min Act: -0.0000

Data Index: 1305716 (C4 (Web Text))

Max Activating Token Index: 237

Text #17

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.6612. Min Act: -0.0000

Data Index: 867308 (C4 (Web Text))

Max Activating Token Index: 500

Text #18

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.6389. Min Act: -0.0000

Data Index: 1381849 (Python Code)

Max Activating Token Index: 797

Text #19

Max Range: 0.8102. Min Range: -0.8102

Max Act: 0.6565. Min Act: -0.0000

Data Index: 1549003 (Python Code)

Max Activating Token Index: 722