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

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

Neuron 5978 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.7640. Min Range: -5.7640

Max Act: 5.7640. Min Act: -0.0000

Data Index: 1379713 (Python Code)

Max Activating Token Index: 697

Text #1

Max Range: 5.7640. Min Range: -5.7640

Max Act: 5.2957. Min Act: -0.0000

Data Index: 1336886 (C4 (Web Text))

Max Activating Token Index: 528

Text #2

Max Range: 5.7640. Min Range: -5.7640

Max Act: 5.7335. Min Act: -0.0000

Data Index: 1442467 (Python Code)

Max Activating Token Index: 303

Text #3

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.6245. Min Act: -0.0000

Data Index: 71344 (C4 (Web Text))

Max Activating Token Index: 53

Text #4

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.8029. Min Act: -0.0000

Data Index: 1513062 (Python Code)

Max Activating Token Index: 46

Text #5

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.3823. Min Act: -0.0000

Data Index: 1161232 (C4 (Web Text))

Max Activating Token Index: 852

Text #6

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.6366. Min Act: -0.0000

Data Index: 1584250 (Python Code)

Max Activating Token Index: 719

Text #7

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.6191. Min Act: -0.0000

Data Index: 657144 (C4 (Web Text))

Max Activating Token Index: 642

Text #8

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.7535. Min Act: -0.0000

Data Index: 1425919 (Python Code)

Max Activating Token Index: 396

Text #9

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.4157. Min Act: -0.0000

Data Index: 198370 (C4 (Web Text))

Max Activating Token Index: 383

Text #10

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.6901. Min Act: -0.0000

Data Index: 1562388 (Python Code)

Max Activating Token Index: 515

Text #11

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.3508. Min Act: -0.0000

Data Index: 1420957 (Python Code)

Max Activating Token Index: 117

Text #12

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.4821. Min Act: -0.0000

Data Index: 1599700 (Python Code)

Max Activating Token Index: 35

Text #13

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.1781. Min Act: -0.0000

Data Index: 1414744 (Python Code)

Max Activating Token Index: 342

Text #14

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.3940. Min Act: -0.0000

Data Index: 1388220 (Python Code)

Max Activating Token Index: 772

Text #15

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.3702. Min Act: -0.0000

Data Index: 1489563 (Python Code)

Max Activating Token Index: 106

Text #16

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.3341. Min Act: -0.0000

Data Index: 1607859 (Python Code)

Max Activating Token Index: 587

Text #17

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.3319. Min Act: -0.0000

Data Index: 1340298 (C4 (Web Text))

Max Activating Token Index: 568

Text #18

Max Range: 5.7640. Min Range: -5.7640

Max Act: 3.9424. Min Act: -0.0000

Data Index: 1260987 (C4 (Web Text))

Max Activating Token Index: 223

Text #19

Max Range: 5.7640. Min Range: -5.7640

Max Act: 4.2489. Min Act: -0.0000

Data Index: 1474320 (Python Code)

Max Activating Token Index: 681