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

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

Neuron 4027 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5536. Min Range: -2.5536

Max Act: 2.5536. Min Act: -0.0000

Data Index: 1602849 (Python Code)

Max Activating Token Index: 200

Text #1

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.4594. Min Act: -0.0000

Data Index: 1471813 (Python Code)

Max Activating Token Index: 230

Text #2

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.5695. Min Act: -0.0000

Data Index: 481553 (C4 (Web Text))

Max Activating Token Index: 866

Text #3

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.5496. Min Act: -0.0000

Data Index: 1278438 (C4 (Web Text))

Max Activating Token Index: 905

Text #4

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.3702. Min Act: -0.0000

Data Index: 559836 (C4 (Web Text))

Max Activating Token Index: 660

Text #5

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.3300. Min Act: -0.0000

Data Index: 760620 (C4 (Web Text))

Max Activating Token Index: 507

Text #6

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.3374. Min Act: -0.0000

Data Index: 376961 (C4 (Web Text))

Max Activating Token Index: 927

Text #7

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.4202. Min Act: -0.0000

Data Index: 2815 (C4 (Web Text))

Max Activating Token Index: 796

Text #8

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.4285. Min Act: -0.0000

Data Index: 708137 (C4 (Web Text))

Max Activating Token Index: 1008

Text #9

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.3872. Min Act: -0.0000

Data Index: 1009632 (C4 (Web Text))

Max Activating Token Index: 888

Text #10

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.3738. Min Act: -0.0000

Data Index: 108107 (C4 (Web Text))

Max Activating Token Index: 842

Text #11

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.4553. Min Act: -0.0000

Data Index: 1236894 (C4 (Web Text))

Max Activating Token Index: 799

Text #12

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.2850. Min Act: -0.0000

Data Index: 1254022 (C4 (Web Text))

Max Activating Token Index: 896

Text #13

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.2762. Min Act: -0.0000

Data Index: 9775 (C4 (Web Text))

Max Activating Token Index: 980

Text #14

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.2604. Min Act: -0.0000

Data Index: 431249 (C4 (Web Text))

Max Activating Token Index: 162

Text #15

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.3114. Min Act: -0.0000

Data Index: 44386 (C4 (Web Text))

Max Activating Token Index: 1001

Text #16

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.2927. Min Act: -0.0000

Data Index: 670391 (C4 (Web Text))

Max Activating Token Index: 978

Text #17

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.2970. Min Act: -0.0000

Data Index: 1359510 (C4 (Web Text))

Max Activating Token Index: 860

Text #18

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.2823. Min Act: -0.0000

Data Index: 739706 (C4 (Web Text))

Max Activating Token Index: 999

Text #19

Max Range: 2.5536. Min Range: -2.5536

Max Act: 1.2443. Min Act: -0.0000

Data Index: 1454027 (Python Code)

Max Activating Token Index: 610