Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 112 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.7608. Min Act: -0.0001

Data Index: 123783 (C4 (Web Text))

Max Activating Token Index: 937

Text #1

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.6891. Min Act: -0.0001

Data Index: 306134 (C4 (Web Text))

Max Activating Token Index: 646

Text #2

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.6586. Min Act: -0.0001

Data Index: 252161 (C4 (Web Text))

Max Activating Token Index: 154

Text #3

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.6585. Min Act: -0.0001

Data Index: 445511 (C4 (Web Text))

Max Activating Token Index: 469

Text #4

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.6654. Min Act: -0.0001

Data Index: 471740 (C4 (Web Text))

Max Activating Token Index: 124

Text #5

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.6063. Min Act: -0.0001

Data Index: 552585 (C4 (Web Text))

Max Activating Token Index: 894

Text #6

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.6257. Min Act: -0.0001

Data Index: 469246 (C4 (Web Text))

Max Activating Token Index: 327

Text #7

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.6092. Min Act: -0.0001

Data Index: 865470 (C4 (Web Text))

Max Activating Token Index: 292

Text #8

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5799. Min Act: -0.0001

Data Index: 1534102 (Python Code)

Max Activating Token Index: 458

Text #9

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5879. Min Act: -0.0001

Data Index: 1360922 (Python Code)

Max Activating Token Index: 776

Text #10

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5709. Min Act: -0.0001

Data Index: 1546856 (Python Code)

Max Activating Token Index: 550

Text #11

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5684. Min Act: -0.0001

Data Index: 349500 (C4 (Web Text))

Max Activating Token Index: 979

Text #12

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5964. Min Act: -0.0001

Data Index: 1451196 (Python Code)

Max Activating Token Index: 464

Text #13

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5811. Min Act: -0.0001

Data Index: 1533495 (Python Code)

Max Activating Token Index: 435

Text #14

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5563. Min Act: -0.0001

Data Index: 107372 (C4 (Web Text))

Max Activating Token Index: 37

Text #15

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5451. Min Act: -0.0001

Data Index: 1368751 (Python Code)

Max Activating Token Index: 272

Text #16

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5680. Min Act: -0.0001

Data Index: 1547952 (Python Code)

Max Activating Token Index: 690

Text #17

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5556. Min Act: -0.0001

Data Index: 1214296 (C4 (Web Text))

Max Activating Token Index: 77

Text #18

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5490. Min Act: -0.0001

Data Index: 308326 (C4 (Web Text))

Max Activating Token Index: 165

Text #19

Max Range: 0.7608. Min Range: -0.7608

Max Act: 0.5079. Min Act: -0.0001

Data Index: 688352 (C4 (Web Text))

Max Activating Token Index: 497