Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2628 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1938. Min Act: -0.0001

Data Index: 1029287 (C4 (Web Text))

Max Activating Token Index: 582

Text #1

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.2340. Min Act: -0.0001

Data Index: 424214 (C4 (Web Text))

Max Activating Token Index: 577

Text #2

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.2028. Min Act: -0.0001

Data Index: 106076 (C4 (Web Text))

Max Activating Token Index: 651

Text #3

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1956. Min Act: -0.0001

Data Index: 825055 (C4 (Web Text))

Max Activating Token Index: 94

Text #4

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.2026. Min Act: -0.0001

Data Index: 269944 (C4 (Web Text))

Max Activating Token Index: 14

Text #5

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.2189. Min Act: -0.0001

Data Index: 321426 (C4 (Web Text))

Max Activating Token Index: 401

Text #6

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1924. Min Act: -0.0001

Data Index: 89104 (C4 (Web Text))

Max Activating Token Index: 353

Text #7

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.2048. Min Act: -0.0001

Data Index: 878773 (C4 (Web Text))

Max Activating Token Index: 729

Text #8

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1941. Min Act: -0.0001

Data Index: 1160105 (C4 (Web Text))

Max Activating Token Index: 90

Text #9

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1994. Min Act: -0.0001

Data Index: 1383841 (Python Code)

Max Activating Token Index: 92

Text #10

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1890. Min Act: -0.0001

Data Index: 288625 (C4 (Web Text))

Max Activating Token Index: 17

Text #11

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1477. Min Act: -0.0001

Data Index: 665275 (C4 (Web Text))

Max Activating Token Index: 871

Text #12

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1843. Min Act: -0.0001

Data Index: 310555 (C4 (Web Text))

Max Activating Token Index: 3

Text #13

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1828. Min Act: -0.0001

Data Index: 1269167 (C4 (Web Text))

Max Activating Token Index: 719

Text #14

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1814. Min Act: -0.0001

Data Index: 609195 (C4 (Web Text))

Max Activating Token Index: 36

Text #15

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1801. Min Act: -0.0001

Data Index: 177210 (C4 (Web Text))

Max Activating Token Index: 520

Text #16

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1803. Min Act: -0.0001

Data Index: 187899 (C4 (Web Text))

Max Activating Token Index: 216

Text #17

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1713. Min Act: -0.0001

Data Index: 1004776 (C4 (Web Text))

Max Activating Token Index: 1014

Text #18

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1744. Min Act: -0.0001

Data Index: 439493 (C4 (Web Text))

Max Activating Token Index: 693

Text #19

Max Range: 0.2340. Min Range: -0.2340

Max Act: 0.1697. Min Act: -0.0001

Data Index: 1115105 (C4 (Web Text))

Max Activating Token Index: 186