Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 740 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.6853. Min Act: -0.0003

Data Index: 427096 (C4 (Web Text))

Max Activating Token Index: 805

Text #1

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5918. Min Act: -0.0003

Data Index: 734557 (C4 (Web Text))

Max Activating Token Index: 672

Text #2

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.6366. Min Act: -0.0003

Data Index: 766792 (C4 (Web Text))

Max Activating Token Index: 956

Text #3

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5849. Min Act: -0.0003

Data Index: 479004 (C4 (Web Text))

Max Activating Token Index: 841

Text #4

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.6025. Min Act: -0.0003

Data Index: 591313 (C4 (Web Text))

Max Activating Token Index: 700

Text #5

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5986. Min Act: -0.0003

Data Index: 771025 (C4 (Web Text))

Max Activating Token Index: 159

Text #6

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5838. Min Act: -0.0003

Data Index: 1223495 (C4 (Web Text))

Max Activating Token Index: 193

Text #7

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.6082. Min Act: -0.0003

Data Index: 981954 (C4 (Web Text))

Max Activating Token Index: 721

Text #8

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5959. Min Act: -0.0003

Data Index: 938367 (C4 (Web Text))

Max Activating Token Index: 132

Text #9

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.6192. Min Act: -0.0003

Data Index: 138644 (C4 (Web Text))

Max Activating Token Index: 795

Text #10

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5209. Min Act: -0.0003

Data Index: 883062 (C4 (Web Text))

Max Activating Token Index: 714

Text #11

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5634. Min Act: -0.0003

Data Index: 762182 (C4 (Web Text))

Max Activating Token Index: 940

Text #12

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5410. Min Act: -0.0004

Data Index: 996040 (C4 (Web Text))

Max Activating Token Index: 903

Text #13

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5750. Min Act: -0.0003

Data Index: 412685 (C4 (Web Text))

Max Activating Token Index: 74

Text #14

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5826. Min Act: -0.0003

Data Index: 217747 (C4 (Web Text))

Max Activating Token Index: 896

Text #15

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5401. Min Act: -0.0003

Data Index: 1076333 (C4 (Web Text))

Max Activating Token Index: 882

Text #16

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5634. Min Act: -0.0003

Data Index: 709079 (C4 (Web Text))

Max Activating Token Index: 199

Text #17

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5536. Min Act: -0.0003

Data Index: 1292272 (C4 (Web Text))

Max Activating Token Index: 296

Text #18

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5509. Min Act: -0.0003

Data Index: 306066 (C4 (Web Text))

Max Activating Token Index: 646

Text #19

Max Range: 0.6853. Min Range: -0.6853

Max Act: 0.5598. Min Act: -0.0003

Data Index: 483499 (C4 (Web Text))

Max Activating Token Index: 597