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

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

Neuron 3287 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: 1.0949. Min Range: -1.0949

Max Act: 1.0949. Min Act: -0.0001

Data Index: 965955 (C4 (Web Text))

Max Activating Token Index: 812

Text #1

Max Range: 1.0949. Min Range: -1.0949

Max Act: 1.0135. Min Act: -0.0001

Data Index: 1287058 (C4 (Web Text))

Max Activating Token Index: 385

Text #2

Max Range: 1.0949. Min Range: -1.0949

Max Act: 1.0508. Min Act: -0.0001

Data Index: 846489 (C4 (Web Text))

Max Activating Token Index: 516

Text #3

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.8989. Min Act: -0.0001

Data Index: 411141 (C4 (Web Text))

Max Activating Token Index: 659

Text #4

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.9830. Min Act: -0.0001

Data Index: 969029 (C4 (Web Text))

Max Activating Token Index: 49

Text #5

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.8759. Min Act: -0.0001

Data Index: 377423 (C4 (Web Text))

Max Activating Token Index: 58

Text #6

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.9643. Min Act: -0.0001

Data Index: 942914 (C4 (Web Text))

Max Activating Token Index: 620

Text #7

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.9626. Min Act: -0.0001

Data Index: 92916 (C4 (Web Text))

Max Activating Token Index: 650

Text #8

Max Range: 1.0949. Min Range: -1.0949

Max Act: 1.0447. Min Act: -0.0001

Data Index: 252845 (C4 (Web Text))

Max Activating Token Index: 891

Text #9

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.8771. Min Act: -0.0001

Data Index: 1518576 (Python Code)

Max Activating Token Index: 21

Text #10

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.8919. Min Act: -0.0001

Data Index: 134819 (C4 (Web Text))

Max Activating Token Index: 967

Text #11

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.9677. Min Act: -0.0001

Data Index: 272697 (C4 (Web Text))

Max Activating Token Index: 422

Text #12

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.9970. Min Act: -0.0001

Data Index: 146274 (C4 (Web Text))

Max Activating Token Index: 37

Text #13

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.9310. Min Act: -0.0001

Data Index: 1321080 (C4 (Web Text))

Max Activating Token Index: 75

Text #14

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.9143. Min Act: -0.0001

Data Index: 531248 (C4 (Web Text))

Max Activating Token Index: 76

Text #15

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.7841. Min Act: -0.0001

Data Index: 1321154 (C4 (Web Text))

Max Activating Token Index: 286

Text #16

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.9172. Min Act: -0.0001

Data Index: 798845 (C4 (Web Text))

Max Activating Token Index: 196

Text #17

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.9287. Min Act: -0.0001

Data Index: 1598039 (Python Code)

Max Activating Token Index: 13

Text #18

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.8761. Min Act: -0.0001

Data Index: 1514474 (Python Code)

Max Activating Token Index: 11

Text #19

Max Range: 1.0949. Min Range: -1.0949

Max Act: 0.8728. Min Act: -0.0001

Data Index: 1501032 (Python Code)

Max Activating Token Index: 12