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

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

Neuron 2616 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0764. Min Act: -0.0001

Data Index: 1354644 (C4 (Web Text))

Max Activating Token Index: 116

Text #1

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0653. Min Act: -0.0001

Data Index: 116982 (C4 (Web Text))

Max Activating Token Index: 947

Text #2

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0700. Min Act: -0.0001

Data Index: 780003 (C4 (Web Text))

Max Activating Token Index: 769

Text #3

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0666. Min Act: -0.0001

Data Index: 116932 (C4 (Web Text))

Max Activating Token Index: 921

Text #4

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0630. Min Act: -0.0001

Data Index: 583492 (C4 (Web Text))

Max Activating Token Index: 725

Text #5

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0588. Min Act: -0.0001

Data Index: 1262424 (C4 (Web Text))

Max Activating Token Index: 531

Text #6

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0580. Min Act: -0.0001

Data Index: 1035479 (C4 (Web Text))

Max Activating Token Index: 164

Text #7

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0596. Min Act: -0.0001

Data Index: 1483914 (Python Code)

Max Activating Token Index: 186

Text #8

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0573. Min Act: -0.0001

Data Index: 637939 (C4 (Web Text))

Max Activating Token Index: 757

Text #9

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0593. Min Act: -0.0001

Data Index: 1343884 (C4 (Web Text))

Max Activating Token Index: 728

Text #10

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0587. Min Act: -0.0001

Data Index: 3881 (C4 (Web Text))

Max Activating Token Index: 908

Text #11

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0575. Min Act: -0.0001

Data Index: 1445934 (Python Code)

Max Activating Token Index: 22

Text #12

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0553. Min Act: -0.0001

Data Index: 903804 (C4 (Web Text))

Max Activating Token Index: 594

Text #13

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0565. Min Act: -0.0001

Data Index: 47703 (C4 (Web Text))

Max Activating Token Index: 695

Text #14

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0563. Min Act: -0.0001

Data Index: 558269 (C4 (Web Text))

Max Activating Token Index: 706

Text #15

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0524. Min Act: -0.0001

Data Index: 349877 (C4 (Web Text))

Max Activating Token Index: 843

Text #16

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0562. Min Act: -0.0001

Data Index: 1558274 (Python Code)

Max Activating Token Index: 173

Text #17

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0559. Min Act: -0.0001

Data Index: 947721 (C4 (Web Text))

Max Activating Token Index: 530

Text #18

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0554. Min Act: -0.0001

Data Index: 19622 (C4 (Web Text))

Max Activating Token Index: 723

Text #19

Max Range: 0.0764. Min Range: -0.0764

Max Act: 0.0557. Min Act: -0.0001

Data Index: 1141870 (C4 (Web Text))

Max Activating Token Index: 426