Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4710 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.1487. Min Act: -0.0001

Data Index: 247917 (C4 (Web Text))

Max Activating Token Index: 368

Text #1

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.1301. Min Act: -0.0001

Data Index: 982748 (C4 (Web Text))

Max Activating Token Index: 781

Text #2

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.1140. Min Act: -0.0001

Data Index: 664560 (C4 (Web Text))

Max Activating Token Index: 400

Text #3

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.1210. Min Act: -0.0001

Data Index: 22702 (C4 (Web Text))

Max Activating Token Index: 74

Text #4

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.1054. Min Act: -0.0001

Data Index: 867797 (C4 (Web Text))

Max Activating Token Index: 237

Text #5

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.1065. Min Act: -0.0001

Data Index: 1408961 (Python Code)

Max Activating Token Index: 17

Text #6

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0972. Min Act: -0.0001

Data Index: 1462798 (Python Code)

Max Activating Token Index: 728

Text #7

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0946. Min Act: -0.0001

Data Index: 1477921 (Python Code)

Max Activating Token Index: 34

Text #8

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0978. Min Act: -0.0001

Data Index: 1602769 (Python Code)

Max Activating Token Index: 340

Text #9

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0935. Min Act: -0.0001

Data Index: 1482826 (Python Code)

Max Activating Token Index: 79

Text #10

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0991. Min Act: -0.0001

Data Index: 1530356 (Python Code)

Max Activating Token Index: 56

Text #11

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0988. Min Act: -0.0001

Data Index: 1483782 (Python Code)

Max Activating Token Index: 698

Text #12

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0969. Min Act: -0.0001

Data Index: 532635 (C4 (Web Text))

Max Activating Token Index: 245

Text #13

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0973. Min Act: -0.0001

Data Index: 1560002 (Python Code)

Max Activating Token Index: 32

Text #14

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0990. Min Act: -0.0001

Data Index: 698483 (C4 (Web Text))

Max Activating Token Index: 695

Text #15

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.1051. Min Act: -0.0001

Data Index: 1374127 (Python Code)

Max Activating Token Index: 264

Text #16

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0919. Min Act: -0.0001

Data Index: 1633973 (Python Code)

Max Activating Token Index: 27

Text #17

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0946. Min Act: -0.0001

Data Index: 1485473 (Python Code)

Max Activating Token Index: 73

Text #18

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0974. Min Act: -0.0001

Data Index: 1436780 (Python Code)

Max Activating Token Index: 693

Text #19

Max Range: 0.1487. Min Range: -0.1487

Max Act: 0.0873. Min Act: -0.0001

Data Index: 1431857 (Python Code)

Max Activating Token Index: 43