Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 249 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.6467. Min Act: -0.0002

Data Index: 625845 (C4 (Web Text))

Max Activating Token Index: 129

Text #1

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5738. Min Act: -0.0002

Data Index: 907969 (C4 (Web Text))

Max Activating Token Index: 529

Text #2

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5478. Min Act: -0.0002

Data Index: 106273 (C4 (Web Text))

Max Activating Token Index: 851

Text #3

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5610. Min Act: -0.0002

Data Index: 829583 (C4 (Web Text))

Max Activating Token Index: 90

Text #4

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5169. Min Act: -0.0002

Data Index: 720151 (C4 (Web Text))

Max Activating Token Index: 217

Text #5

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5144. Min Act: -0.0002

Data Index: 309369 (C4 (Web Text))

Max Activating Token Index: 105

Text #6

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5001. Min Act: -0.0002

Data Index: 1059154 (C4 (Web Text))

Max Activating Token Index: 110

Text #7

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5027. Min Act: -0.0002

Data Index: 1011830 (C4 (Web Text))

Max Activating Token Index: 732

Text #8

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5269. Min Act: -0.0002

Data Index: 455242 (C4 (Web Text))

Max Activating Token Index: 560

Text #9

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5353. Min Act: -0.0002

Data Index: 1047949 (C4 (Web Text))

Max Activating Token Index: 587

Text #10

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5029. Min Act: -0.0002

Data Index: 456243 (C4 (Web Text))

Max Activating Token Index: 883

Text #11

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.4734. Min Act: -0.0002

Data Index: 1144440 (C4 (Web Text))

Max Activating Token Index: 930

Text #12

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.5088. Min Act: -0.0002

Data Index: 901699 (C4 (Web Text))

Max Activating Token Index: 613

Text #13

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.4823. Min Act: -0.0002

Data Index: 1096671 (C4 (Web Text))

Max Activating Token Index: 958

Text #14

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.4943. Min Act: -0.0002

Data Index: 252813 (C4 (Web Text))

Max Activating Token Index: 904

Text #15

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.4813. Min Act: -0.0002

Data Index: 1262897 (C4 (Web Text))

Max Activating Token Index: 680

Text #16

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.4778. Min Act: -0.0002

Data Index: 646592 (C4 (Web Text))

Max Activating Token Index: 522

Text #17

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.4749. Min Act: -0.0002

Data Index: 639356 (C4 (Web Text))

Max Activating Token Index: 481

Text #18

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.4564. Min Act: -0.0002

Data Index: 10636 (C4 (Web Text))

Max Activating Token Index: 686

Text #19

Max Range: 0.6467. Min Range: -0.6467

Max Act: 0.4841. Min Act: -0.0002

Data Index: 84890 (C4 (Web Text))

Max Activating Token Index: 786