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

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

Neuron 2128 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.1112. Min Act: -0.0001

Data Index: 87903 (C4 (Web Text))

Max Activating Token Index: 434

Text #1

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.1099. Min Act: -0.0001

Data Index: 899935 (C4 (Web Text))

Max Activating Token Index: 142

Text #2

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.1001. Min Act: -0.0000

Data Index: 384571 (C4 (Web Text))

Max Activating Token Index: 446

Text #3

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0917. Min Act: -0.0000

Data Index: 4222 (C4 (Web Text))

Max Activating Token Index: 914

Text #4

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0951. Min Act: -0.0001

Data Index: 378433 (C4 (Web Text))

Max Activating Token Index: 250

Text #5

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0946. Min Act: -0.0001

Data Index: 565296 (C4 (Web Text))

Max Activating Token Index: 711

Text #6

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0862. Min Act: -0.0001

Data Index: 970726 (C4 (Web Text))

Max Activating Token Index: 249

Text #7

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0886. Min Act: -0.0001

Data Index: 616684 (C4 (Web Text))

Max Activating Token Index: 107

Text #8

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0904. Min Act: -0.0001

Data Index: 1326072 (C4 (Web Text))

Max Activating Token Index: 292

Text #9

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0833. Min Act: -0.0001

Data Index: 670357 (C4 (Web Text))

Max Activating Token Index: 945

Text #10

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0836. Min Act: -0.0001

Data Index: 564870 (C4 (Web Text))

Max Activating Token Index: 267

Text #11

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0854. Min Act: -0.0000

Data Index: 583947 (C4 (Web Text))

Max Activating Token Index: 983

Text #12

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0876. Min Act: -0.0001

Data Index: 1527759 (Python Code)

Max Activating Token Index: 54

Text #13

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0847. Min Act: -0.0001

Data Index: 590999 (C4 (Web Text))

Max Activating Token Index: 571

Text #14

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0788. Min Act: -0.0000

Data Index: 734963 (C4 (Web Text))

Max Activating Token Index: 674

Text #15

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0839. Min Act: -0.0001

Data Index: 202439 (C4 (Web Text))

Max Activating Token Index: 448

Text #16

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0834. Min Act: -0.0001

Data Index: 1042307 (C4 (Web Text))

Max Activating Token Index: 824

Text #17

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0825. Min Act: -0.0000

Data Index: 721296 (C4 (Web Text))

Max Activating Token Index: 108

Text #18

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0847. Min Act: -0.0001

Data Index: 966537 (C4 (Web Text))

Max Activating Token Index: 808

Text #19

Max Range: 0.1112. Min Range: -0.1112

Max Act: 0.0812. Min Act: -0.0001

Data Index: 593508 (C4 (Web Text))

Max Activating Token Index: 899