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

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

Neuron 4307 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0485. Min Range: -1.0485

Max Act: 1.0485. Min Act: -0.0001

Data Index: 1337814 (C4 (Web Text))

Max Activating Token Index: 983

Text #1

Max Range: 1.0485. Min Range: -1.0485

Max Act: 1.0234. Min Act: -0.0001

Data Index: 1184745 (C4 (Web Text))

Max Activating Token Index: 931

Text #2

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9546. Min Act: -0.0001

Data Index: 941293 (C4 (Web Text))

Max Activating Token Index: 1005

Text #3

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9851. Min Act: -0.0001

Data Index: 1021858 (C4 (Web Text))

Max Activating Token Index: 824

Text #4

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9520. Min Act: -0.0001

Data Index: 1648088 (Python Code)

Max Activating Token Index: 6

Text #5

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9481. Min Act: -0.0001

Data Index: 1108781 (C4 (Web Text))

Max Activating Token Index: 96

Text #6

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9374. Min Act: -0.0001

Data Index: 1303094 (C4 (Web Text))

Max Activating Token Index: 1005

Text #7

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9064. Min Act: -0.0001

Data Index: 1338984 (C4 (Web Text))

Max Activating Token Index: 802

Text #8

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9830. Min Act: -0.0001

Data Index: 823200 (C4 (Web Text))

Max Activating Token Index: 941

Text #9

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.8943. Min Act: -0.0001

Data Index: 1126345 (C4 (Web Text))

Max Activating Token Index: 753

Text #10

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9021. Min Act: -0.0001

Data Index: 320957 (C4 (Web Text))

Max Activating Token Index: 463

Text #11

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9474. Min Act: -0.0001

Data Index: 1354390 (C4 (Web Text))

Max Activating Token Index: 587

Text #12

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.8607. Min Act: -0.0001

Data Index: 29691 (C4 (Web Text))

Max Activating Token Index: 488

Text #13

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9373. Min Act: -0.0001

Data Index: 182587 (C4 (Web Text))

Max Activating Token Index: 536

Text #14

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.8556. Min Act: -0.0001

Data Index: 28057 (C4 (Web Text))

Max Activating Token Index: 655

Text #15

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9124. Min Act: -0.0001

Data Index: 679207 (C4 (Web Text))

Max Activating Token Index: 130

Text #16

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9262. Min Act: -0.0001

Data Index: 1640161 (Python Code)

Max Activating Token Index: 716

Text #17

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.9315. Min Act: -0.0001

Data Index: 248757 (C4 (Web Text))

Max Activating Token Index: 790

Text #18

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.8630. Min Act: -0.0001

Data Index: 422161 (C4 (Web Text))

Max Activating Token Index: 518

Text #19

Max Range: 1.0485. Min Range: -1.0485

Max Act: 0.8914. Min Act: -0.0001

Data Index: 468886 (C4 (Web Text))

Max Activating Token Index: 659