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

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

Neuron 786 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: 0.0910. Min Range: -0.0910

Max Act: 0.0910. Min Act: -0.0001

Data Index: 549688 (C4 (Web Text))

Max Activating Token Index: 981

Text #1

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0798. Min Act: -0.0001

Data Index: 610493 (C4 (Web Text))

Max Activating Token Index: 597

Text #2

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0753. Min Act: -0.0001

Data Index: 1015438 (C4 (Web Text))

Max Activating Token Index: 717

Text #3

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0698. Min Act: -0.0001

Data Index: 332899 (C4 (Web Text))

Max Activating Token Index: 984

Text #4

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0699. Min Act: -0.0001

Data Index: 148311 (C4 (Web Text))

Max Activating Token Index: 90

Text #5

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0663. Min Act: -0.0001

Data Index: 318321 (C4 (Web Text))

Max Activating Token Index: 735

Text #6

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0698. Min Act: -0.0001

Data Index: 462988 (C4 (Web Text))

Max Activating Token Index: 800

Text #7

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0671. Min Act: -0.0001

Data Index: 110713 (C4 (Web Text))

Max Activating Token Index: 922

Text #8

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0697. Min Act: -0.0001

Data Index: 1372281 (Python Code)

Max Activating Token Index: 474

Text #9

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0669. Min Act: -0.0001

Data Index: 88175 (C4 (Web Text))

Max Activating Token Index: 719

Text #10

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0640. Min Act: -0.0001

Data Index: 974412 (C4 (Web Text))

Max Activating Token Index: 604

Text #11

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0643. Min Act: -0.0001

Data Index: 900573 (C4 (Web Text))

Max Activating Token Index: 970

Text #12

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0654. Min Act: -0.0001

Data Index: 879295 (C4 (Web Text))

Max Activating Token Index: 835

Text #13

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0600. Min Act: -0.0001

Data Index: 844208 (C4 (Web Text))

Max Activating Token Index: 181

Text #14

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0623. Min Act: -0.0001

Data Index: 1058752 (C4 (Web Text))

Max Activating Token Index: 692

Text #15

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0632. Min Act: -0.0001

Data Index: 239793 (C4 (Web Text))

Max Activating Token Index: 795

Text #16

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0601. Min Act: -0.0001

Data Index: 588108 (C4 (Web Text))

Max Activating Token Index: 936

Text #17

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0558. Min Act: -0.0001

Data Index: 768984 (C4 (Web Text))

Max Activating Token Index: 38

Text #18

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0556. Min Act: -0.0001

Data Index: 450059 (C4 (Web Text))

Max Activating Token Index: 634

Text #19

Max Range: 0.0910. Min Range: -0.0910

Max Act: 0.0550. Min Act: -0.0001

Data Index: 1064896 (C4 (Web Text))

Max Activating Token Index: 513