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

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

Neuron 3007 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.2029. Min Act: -0.0001

Data Index: 584319 (C4 (Web Text))

Max Activating Token Index: 165

Text #1

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.2081. Min Act: -0.0001

Data Index: 742828 (C4 (Web Text))

Max Activating Token Index: 463

Text #2

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1890. Min Act: -0.0001

Data Index: 1119772 (C4 (Web Text))

Max Activating Token Index: 463

Text #3

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1929. Min Act: -0.0001

Data Index: 1535689 (Python Code)

Max Activating Token Index: 5

Text #4

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1820. Min Act: -0.0001

Data Index: 1534946 (Python Code)

Max Activating Token Index: 662

Text #5

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1857. Min Act: -0.0001

Data Index: 1136672 (C4 (Web Text))

Max Activating Token Index: 154

Text #6

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1749. Min Act: -0.0001

Data Index: 355962 (C4 (Web Text))

Max Activating Token Index: 749

Text #7

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1698. Min Act: -0.0001

Data Index: 474220 (C4 (Web Text))

Max Activating Token Index: 234

Text #8

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1742. Min Act: -0.0001

Data Index: 288333 (C4 (Web Text))

Max Activating Token Index: 452

Text #9

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1686. Min Act: -0.0001

Data Index: 500961 (C4 (Web Text))

Max Activating Token Index: 619

Text #10

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1668. Min Act: -0.0001

Data Index: 1480959 (Python Code)

Max Activating Token Index: 22

Text #11

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1660. Min Act: -0.0001

Data Index: 1366847 (Python Code)

Max Activating Token Index: 10

Text #12

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1586. Min Act: -0.0001

Data Index: 1516210 (Python Code)

Max Activating Token Index: 586

Text #13

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1648. Min Act: -0.0001

Data Index: 453997 (C4 (Web Text))

Max Activating Token Index: 682

Text #14

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1621. Min Act: -0.0001

Data Index: 1393510 (Python Code)

Max Activating Token Index: 822

Text #15

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1578. Min Act: -0.0001

Data Index: 934172 (C4 (Web Text))

Max Activating Token Index: 571

Text #16

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1647. Min Act: -0.0001

Data Index: 1587623 (Python Code)

Max Activating Token Index: 5

Text #17

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1634. Min Act: -0.0001

Data Index: 586128 (C4 (Web Text))

Max Activating Token Index: 728

Text #18

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1577. Min Act: -0.0001

Data Index: 1587921 (Python Code)

Max Activating Token Index: 918

Text #19

Max Range: 0.2081. Min Range: -0.2081

Max Act: 0.1501. Min Act: -0.0001

Data Index: 201702 (C4 (Web Text))

Max Activating Token Index: 142