Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2816 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0903. Min Act: -0.0001

Data Index: 366862 (C4 (Web Text))

Max Activating Token Index: 351

Text #1

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0863. Min Act: -0.0001

Data Index: 270832 (C4 (Web Text))

Max Activating Token Index: 100

Text #2

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0954. Min Act: -0.0001

Data Index: 152030 (C4 (Web Text))

Max Activating Token Index: 928

Text #3

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0858. Min Act: -0.0001

Data Index: 1183084 (C4 (Web Text))

Max Activating Token Index: 495

Text #4

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0800. Min Act: -0.0001

Data Index: 287592 (C4 (Web Text))

Max Activating Token Index: 726

Text #5

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0869. Min Act: -0.0001

Data Index: 837566 (C4 (Web Text))

Max Activating Token Index: 596

Text #6

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0719. Min Act: -0.0001

Data Index: 1575536 (Python Code)

Max Activating Token Index: 418

Text #7

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0973. Min Act: -0.0001

Data Index: 822710 (C4 (Web Text))

Max Activating Token Index: 813

Text #8

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0907. Min Act: -0.0001

Data Index: 362093 (C4 (Web Text))

Max Activating Token Index: 551

Text #9

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0824. Min Act: -0.0001

Data Index: 994830 (C4 (Web Text))

Max Activating Token Index: 32

Text #10

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0678. Min Act: -0.0001

Data Index: 1393251 (Python Code)

Max Activating Token Index: 891

Text #11

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0746. Min Act: -0.0001

Data Index: 1136892 (C4 (Web Text))

Max Activating Token Index: 873

Text #12

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0751. Min Act: -0.0001

Data Index: 961436 (C4 (Web Text))

Max Activating Token Index: 455

Text #13

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0721. Min Act: -0.0001

Data Index: 1531159 (Python Code)

Max Activating Token Index: 77

Text #14

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0761. Min Act: -0.0001

Data Index: 1011858 (C4 (Web Text))

Max Activating Token Index: 11

Text #15

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0729. Min Act: -0.0001

Data Index: 1522965 (Python Code)

Max Activating Token Index: 558

Text #16

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0677. Min Act: -0.0001

Data Index: 998757 (C4 (Web Text))

Max Activating Token Index: 888

Text #17

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0724. Min Act: -0.0001

Data Index: 245904 (C4 (Web Text))

Max Activating Token Index: 947

Text #18

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0711. Min Act: -0.0001

Data Index: 1498098 (Python Code)

Max Activating Token Index: 33

Text #19

Max Range: 0.0973. Min Range: -0.0973

Max Act: 0.0759. Min Act: -0.0001

Data Index: 1449107 (Python Code)

Max Activating Token Index: 539