Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 146 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.1401. Min Act: -0.0001

Data Index: 1595128 (Python Code)

Max Activating Token Index: 831

Text #1

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0945. Min Act: -0.0001

Data Index: 1175015 (C4 (Web Text))

Max Activating Token Index: 94

Text #2

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0871. Min Act: -0.0001

Data Index: 697425 (C4 (Web Text))

Max Activating Token Index: 185

Text #3

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0900. Min Act: -0.0001

Data Index: 900159 (C4 (Web Text))

Max Activating Token Index: 331

Text #4

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0885. Min Act: -0.0001

Data Index: 1546500 (Python Code)

Max Activating Token Index: 269

Text #5

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0880. Min Act: -0.0001

Data Index: 105906 (C4 (Web Text))

Max Activating Token Index: 581

Text #6

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0811. Min Act: -0.0001

Data Index: 272803 (C4 (Web Text))

Max Activating Token Index: 722

Text #7

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0842. Min Act: -0.0001

Data Index: 367023 (C4 (Web Text))

Max Activating Token Index: 214

Text #8

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0818. Min Act: -0.0001

Data Index: 503033 (C4 (Web Text))

Max Activating Token Index: 814

Text #9

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0730. Min Act: -0.0001

Data Index: 1085858 (C4 (Web Text))

Max Activating Token Index: 871

Text #10

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0790. Min Act: -0.0001

Data Index: 1037245 (C4 (Web Text))

Max Activating Token Index: 179

Text #11

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0821. Min Act: -0.0001

Data Index: 301749 (C4 (Web Text))

Max Activating Token Index: 566

Text #12

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0812. Min Act: -0.0001

Data Index: 1040621 (C4 (Web Text))

Max Activating Token Index: 78

Text #13

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0802. Min Act: -0.0001

Data Index: 479534 (C4 (Web Text))

Max Activating Token Index: 654

Text #14

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0730. Min Act: -0.0001

Data Index: 839211 (C4 (Web Text))

Max Activating Token Index: 433

Text #15

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0709. Min Act: -0.0001

Data Index: 1259527 (C4 (Web Text))

Max Activating Token Index: 665

Text #16

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0716. Min Act: -0.0001

Data Index: 160901 (C4 (Web Text))

Max Activating Token Index: 54

Text #17

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0763. Min Act: -0.0001

Data Index: 547055 (C4 (Web Text))

Max Activating Token Index: 367

Text #18

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0718. Min Act: -0.0001

Data Index: 1161286 (C4 (Web Text))

Max Activating Token Index: 394

Text #19

Max Range: 0.1401. Min Range: -0.1401

Max Act: 0.0726. Min Act: -0.0001

Data Index: 1647871 (Python Code)

Max Activating Token Index: 498