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

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

Neuron 1445 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5762. Min Range: -2.5762

Max Act: 2.5762. Min Act: -0.0000

Data Index: 459113 (C4 (Web Text))

Max Activating Token Index: 873

Text #1

Max Range: 2.5762. Min Range: -2.5762

Max Act: 2.0692. Min Act: -0.0000

Data Index: 662619 (C4 (Web Text))

Max Activating Token Index: 513

Text #2

Max Range: 2.5762. Min Range: -2.5762

Max Act: 2.0869. Min Act: -0.0000

Data Index: 1185682 (C4 (Web Text))

Max Activating Token Index: 515

Text #3

Max Range: 2.5762. Min Range: -2.5762

Max Act: 2.0283. Min Act: -0.0000

Data Index: 340022 (C4 (Web Text))

Max Activating Token Index: 683

Text #4

Max Range: 2.5762. Min Range: -2.5762

Max Act: 2.0604. Min Act: -0.0000

Data Index: 23143 (C4 (Web Text))

Max Activating Token Index: 240

Text #5

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.8807. Min Act: -0.0000

Data Index: 1580739 (Python Code)

Max Activating Token Index: 568

Text #6

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.9861. Min Act: -0.0000

Data Index: 1648375 (Python Code)

Max Activating Token Index: 469

Text #7

Max Range: 2.5762. Min Range: -2.5762

Max Act: 2.1071. Min Act: -0.0000

Data Index: 597756 (C4 (Web Text))

Max Activating Token Index: 54

Text #8

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.9781. Min Act: -0.0000

Data Index: 633156 (C4 (Web Text))

Max Activating Token Index: 346

Text #9

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.7598. Min Act: -0.0000

Data Index: 610443 (C4 (Web Text))

Max Activating Token Index: 76

Text #10

Max Range: 2.5762. Min Range: -2.5762

Max Act: 2.0760. Min Act: -0.0000

Data Index: 228771 (C4 (Web Text))

Max Activating Token Index: 213

Text #11

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.8615. Min Act: -0.0000

Data Index: 1628005 (Python Code)

Max Activating Token Index: 887

Text #12

Max Range: 2.5762. Min Range: -2.5762

Max Act: 2.1025. Min Act: -0.0000

Data Index: 1199095 (C4 (Web Text))

Max Activating Token Index: 278

Text #13

Max Range: 2.5762. Min Range: -2.5762

Max Act: 2.0417. Min Act: -0.0000

Data Index: 878058 (C4 (Web Text))

Max Activating Token Index: 248

Text #14

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.9195. Min Act: -0.0000

Data Index: 385873 (C4 (Web Text))

Max Activating Token Index: 632

Text #15

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.9013. Min Act: -0.0000

Data Index: 377523 (C4 (Web Text))

Max Activating Token Index: 471

Text #16

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.4715. Min Act: -0.0000

Data Index: 668049 (C4 (Web Text))

Max Activating Token Index: 747

Text #17

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.8044. Min Act: -0.0000

Data Index: 1494332 (Python Code)

Max Activating Token Index: 540

Text #18

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.8847. Min Act: -0.0000

Data Index: 662790 (C4 (Web Text))

Max Activating Token Index: 545

Text #19

Max Range: 2.5762. Min Range: -2.5762

Max Act: 1.7789. Min Act: -0.0000

Data Index: 58345 (C4 (Web Text))

Max Activating Token Index: 618