Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 586 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.4784. Min Range: -4.4784

Max Act: 4.4270. Min Act: -0.0000

Data Index: 1022656 (C4 (Web Text))

Max Activating Token Index: 542

Text #1

Max Range: 4.4784. Min Range: -4.4784

Max Act: 4.4784. Min Act: -0.0000

Data Index: 313006 (C4 (Web Text))

Max Activating Token Index: 394

Text #2

Max Range: 4.4784. Min Range: -4.4784

Max Act: 3.8872. Min Act: -0.0000

Data Index: 1537761 (Python Code)

Max Activating Token Index: 322

Text #3

Max Range: 4.4784. Min Range: -4.4784

Max Act: 4.1604. Min Act: -0.0000

Data Index: 38095 (C4 (Web Text))

Max Activating Token Index: 33

Text #4

Max Range: 4.4784. Min Range: -4.4784

Max Act: 3.1082. Min Act: -0.0000

Data Index: 758513 (C4 (Web Text))

Max Activating Token Index: 850

Text #5

Max Range: 4.4784. Min Range: -4.4784

Max Act: 3.5884. Min Act: -0.0000

Data Index: 1084265 (C4 (Web Text))

Max Activating Token Index: 695

Text #6

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.8559. Min Act: -0.0000

Data Index: 428635 (C4 (Web Text))

Max Activating Token Index: 1007

Text #7

Max Range: 4.4784. Min Range: -4.4784

Max Act: 3.1055. Min Act: -0.0000

Data Index: 1566902 (Python Code)

Max Activating Token Index: 44

Text #8

Max Range: 4.4784. Min Range: -4.4784

Max Act: 3.1466. Min Act: -0.0000

Data Index: 1181173 (C4 (Web Text))

Max Activating Token Index: 690

Text #9

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.5545. Min Act: -0.0000

Data Index: 1485742 (Python Code)

Max Activating Token Index: 211

Text #10

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.6428. Min Act: -0.0000

Data Index: 1394115 (Python Code)

Max Activating Token Index: 39

Text #11

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.7416. Min Act: -0.0000

Data Index: 1405790 (Python Code)

Max Activating Token Index: 474

Text #12

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.7876. Min Act: -0.0000

Data Index: 1377381 (Python Code)

Max Activating Token Index: 679

Text #13

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.7733. Min Act: -0.0000

Data Index: 235232 (C4 (Web Text))

Max Activating Token Index: 358

Text #14

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.2218. Min Act: -0.0000

Data Index: 1651082 (Python Code)

Max Activating Token Index: 190

Text #15

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.5208. Min Act: -0.0000

Data Index: 1591259 (Python Code)

Max Activating Token Index: 410

Text #16

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.1622. Min Act: -0.0000

Data Index: 1423115 (Python Code)

Max Activating Token Index: 49

Text #17

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.5156. Min Act: -0.0000

Data Index: 1566953 (Python Code)

Max Activating Token Index: 75

Text #18

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.4283. Min Act: -0.0000

Data Index: 1537046 (Python Code)

Max Activating Token Index: 862

Text #19

Max Range: 4.4784. Min Range: -4.4784

Max Act: 2.3836. Min Act: -0.0000

Data Index: 1564425 (Python Code)

Max Activating Token Index: 75