Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 203 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0630. Min Act: -0.0002

Data Index: 39303 (C4 (Web Text))

Max Activating Token Index: 640

Text #1

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0638. Min Act: -0.0002

Data Index: 817548 (C4 (Web Text))

Max Activating Token Index: 542

Text #2

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0596. Min Act: -0.0002

Data Index: 97186 (C4 (Web Text))

Max Activating Token Index: 908

Text #3

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0693. Min Act: -0.0002

Data Index: 735106 (C4 (Web Text))

Max Activating Token Index: 991

Text #4

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0616. Min Act: -0.0002

Data Index: 472847 (C4 (Web Text))

Max Activating Token Index: 488

Text #5

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0528. Min Act: -0.0002

Data Index: 1236152 (C4 (Web Text))

Max Activating Token Index: 167

Text #6

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0591. Min Act: -0.0002

Data Index: 24087 (C4 (Web Text))

Max Activating Token Index: 927

Text #7

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0550. Min Act: -0.0002

Data Index: 370826 (C4 (Web Text))

Max Activating Token Index: 610

Text #8

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0539. Min Act: -0.0002

Data Index: 767773 (C4 (Web Text))

Max Activating Token Index: 602

Text #9

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0550. Min Act: -0.0002

Data Index: 168853 (C4 (Web Text))

Max Activating Token Index: 552

Text #10

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0525. Min Act: -0.0002

Data Index: 674539 (C4 (Web Text))

Max Activating Token Index: 77

Text #11

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0559. Min Act: -0.0002

Data Index: 1157787 (C4 (Web Text))

Max Activating Token Index: 728

Text #12

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0519. Min Act: -0.0002

Data Index: 1026754 (C4 (Web Text))

Max Activating Token Index: 310

Text #13

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0526. Min Act: -0.0002

Data Index: 953709 (C4 (Web Text))

Max Activating Token Index: 234

Text #14

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0531. Min Act: -0.0002

Data Index: 80698 (C4 (Web Text))

Max Activating Token Index: 111

Text #15

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0515. Min Act: -0.0002

Data Index: 1525445 (Python Code)

Max Activating Token Index: 404

Text #16

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0505. Min Act: -0.0002

Data Index: 188200 (C4 (Web Text))

Max Activating Token Index: 807

Text #17

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0501. Min Act: -0.0002

Data Index: 1267451 (C4 (Web Text))

Max Activating Token Index: 174

Text #18

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0495. Min Act: -0.0002

Data Index: 1507056 (Python Code)

Max Activating Token Index: 566

Text #19

Max Range: 0.0693. Min Range: -0.0693

Max Act: 0.0506. Min Act: -0.0002

Data Index: 977013 (C4 (Web Text))

Max Activating Token Index: 841