Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3900 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.2290. Min Act: -0.0001

Data Index: 274881 (C4 (Web Text))

Max Activating Token Index: 29

Text #1

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.2184. Min Act: -0.0000

Data Index: 799237 (C4 (Web Text))

Max Activating Token Index: 554

Text #2

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.2147. Min Act: -0.0000

Data Index: 948937 (C4 (Web Text))

Max Activating Token Index: 31

Text #3

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1999. Min Act: -0.0001

Data Index: 973738 (C4 (Web Text))

Max Activating Token Index: 960

Text #4

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.2027. Min Act: -0.0001

Data Index: 1459284 (Python Code)

Max Activating Token Index: 529

Text #5

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.2045. Min Act: -0.0001

Data Index: 91196 (C4 (Web Text))

Max Activating Token Index: 803

Text #6

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1913. Min Act: -0.0000

Data Index: 1232178 (C4 (Web Text))

Max Activating Token Index: 299

Text #7

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1926. Min Act: -0.0000

Data Index: 624141 (C4 (Web Text))

Max Activating Token Index: 25

Text #8

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1966. Min Act: -0.0001

Data Index: 1058786 (C4 (Web Text))

Max Activating Token Index: 324

Text #9

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1903. Min Act: -0.0001

Data Index: 156021 (C4 (Web Text))

Max Activating Token Index: 682

Text #10

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1824. Min Act: -0.0001

Data Index: 1022244 (C4 (Web Text))

Max Activating Token Index: 54

Text #11

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1811. Min Act: -0.0001

Data Index: 930733 (C4 (Web Text))

Max Activating Token Index: 473

Text #12

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1604. Min Act: -0.0001

Data Index: 760230 (C4 (Web Text))

Max Activating Token Index: 34

Text #13

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1711. Min Act: -0.0000

Data Index: 1024925 (C4 (Web Text))

Max Activating Token Index: 1020

Text #14

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1774. Min Act: -0.0000

Data Index: 861066 (C4 (Web Text))

Max Activating Token Index: 432

Text #15

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1727. Min Act: -0.0000

Data Index: 1152405 (C4 (Web Text))

Max Activating Token Index: 18

Text #16

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1761. Min Act: -0.0001

Data Index: 1580415 (Python Code)

Max Activating Token Index: 376

Text #17

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1744. Min Act: -0.0001

Data Index: 1581661 (Python Code)

Max Activating Token Index: 609

Text #18

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1739. Min Act: -0.0000

Data Index: 1236560 (C4 (Web Text))

Max Activating Token Index: 429

Text #19

Max Range: 0.2290. Min Range: -0.2290

Max Act: 0.1686. Min Act: -0.0001

Data Index: 695560 (C4 (Web Text))

Max Activating Token Index: 49