Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2036 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0772. Min Act: -0.0003

Data Index: 1446045 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0780. Min Act: -0.0003

Data Index: 760129 (C4 (Web Text))

Max Activating Token Index: 991

Text #2

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0769. Min Act: -0.0003

Data Index: 917085 (C4 (Web Text))

Max Activating Token Index: 657

Text #3

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0741. Min Act: -0.0003

Data Index: 224525 (C4 (Web Text))

Max Activating Token Index: 76

Text #4

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0713. Min Act: -0.0003

Data Index: 700506 (C4 (Web Text))

Max Activating Token Index: 54

Text #5

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0741. Min Act: -0.0003

Data Index: 495789 (C4 (Web Text))

Max Activating Token Index: 54

Text #6

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0772. Min Act: -0.0003

Data Index: 810243 (C4 (Web Text))

Max Activating Token Index: 926

Text #7

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0723. Min Act: -0.0003

Data Index: 1070026 (C4 (Web Text))

Max Activating Token Index: 289

Text #8

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0722. Min Act: -0.0003

Data Index: 160167 (C4 (Web Text))

Max Activating Token Index: 54

Text #9

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0722. Min Act: -0.0003

Data Index: 839109 (C4 (Web Text))

Max Activating Token Index: 316

Text #10

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0707. Min Act: -0.0003

Data Index: 1097706 (C4 (Web Text))

Max Activating Token Index: 94

Text #11

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0694. Min Act: -0.0003

Data Index: 789512 (C4 (Web Text))

Max Activating Token Index: 153

Text #12

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0719. Min Act: -0.0003

Data Index: 1216657 (C4 (Web Text))

Max Activating Token Index: 31

Text #13

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0693. Min Act: -0.0003

Data Index: 253739 (C4 (Web Text))

Max Activating Token Index: 991

Text #14

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0707. Min Act: -0.0003

Data Index: 923102 (C4 (Web Text))

Max Activating Token Index: 38

Text #15

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0699. Min Act: -0.0003

Data Index: 1051519 (C4 (Web Text))

Max Activating Token Index: 906

Text #16

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0664. Min Act: -0.0003

Data Index: 675064 (C4 (Web Text))

Max Activating Token Index: 393

Text #17

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0669. Min Act: -0.0003

Data Index: 532786 (C4 (Web Text))

Max Activating Token Index: 339

Text #18

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0701. Min Act: -0.0003

Data Index: 231287 (C4 (Web Text))

Max Activating Token Index: 22

Text #19

Max Range: 0.0780. Min Range: -0.0780

Max Act: 0.0699. Min Act: -0.0003

Data Index: 1333080 (C4 (Web Text))

Max Activating Token Index: 138