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

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

Neuron 2815 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.3753. Min Act: -0.0000

Data Index: 870783 (C4 (Web Text))

Max Activating Token Index: 420

Text #1

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.1795. Min Act: -0.0000

Data Index: 1454368 (Python Code)

Max Activating Token Index: 1001

Text #2

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.2154. Min Act: -0.0000

Data Index: 1522502 (Python Code)

Max Activating Token Index: 882

Text #3

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.2961. Min Act: -0.0001

Data Index: 204312 (C4 (Web Text))

Max Activating Token Index: 33

Text #4

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.2215. Min Act: -0.0001

Data Index: 1512293 (Python Code)

Max Activating Token Index: 965

Text #5

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.1919. Min Act: -0.0001

Data Index: 1405557 (Python Code)

Max Activating Token Index: 934

Text #6

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.2872. Min Act: -0.0001

Data Index: 1422709 (Python Code)

Max Activating Token Index: 1013

Text #7

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.0506. Min Act: -0.0001

Data Index: 6870 (C4 (Web Text))

Max Activating Token Index: 721

Text #8

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.1087. Min Act: -0.0000

Data Index: 1369026 (Python Code)

Max Activating Token Index: 887

Text #9

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.1330. Min Act: -0.0001

Data Index: 610381 (C4 (Web Text))

Max Activating Token Index: 236

Text #10

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.0032. Min Act: -0.0000

Data Index: 1375250 (Python Code)

Max Activating Token Index: 235

Text #11

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.0011. Min Act: -0.0000

Data Index: 1370588 (Python Code)

Max Activating Token Index: 667

Text #12

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.1269. Min Act: -0.0001

Data Index: 1512532 (Python Code)

Max Activating Token Index: 1005

Text #13

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.1240. Min Act: -0.0001

Data Index: 1590177 (Python Code)

Max Activating Token Index: 973

Text #14

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.0636. Min Act: -0.0000

Data Index: 1598191 (Python Code)

Max Activating Token Index: 690

Text #15

Max Range: 1.3753. Min Range: -1.3753

Max Act: 0.9611. Min Act: -0.0001

Data Index: 1427020 (Python Code)

Max Activating Token Index: 761

Text #16

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.1952. Min Act: -0.0001

Data Index: 1483486 (Python Code)

Max Activating Token Index: 97

Text #17

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.0867. Min Act: -0.0000

Data Index: 930107 (C4 (Web Text))

Max Activating Token Index: 921

Text #18

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.0611. Min Act: -0.0001

Data Index: 298006 (C4 (Web Text))

Max Activating Token Index: 395

Text #19

Max Range: 1.3753. Min Range: -1.3753

Max Act: 1.0063. Min Act: -0.0000

Data Index: 1618455 (Python Code)

Max Activating Token Index: 198