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

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

Neuron 4532 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.3540. Min Range: -2.3540

Max Act: 2.3099. Min Act: 0.0000

Data Index: 873579 (C4 (Web Text))

Max Activating Token Index: 115

Text #1

Max Range: 2.3540. Min Range: -2.3540

Max Act: 2.3540. Min Act: -0.0000

Data Index: 1433118 (Python Code)

Max Activating Token Index: 58

Text #2

Max Range: 2.3540. Min Range: -2.3540

Max Act: 2.0138. Min Act: -0.0000

Data Index: 1390592 (Python Code)

Max Activating Token Index: 386

Text #3

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.9162. Min Act: 0.0000

Data Index: 965506 (C4 (Web Text))

Max Activating Token Index: 789

Text #4

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.8248. Min Act: -0.0000

Data Index: 1636545 (Python Code)

Max Activating Token Index: 957

Text #5

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.7955. Min Act: -0.0000

Data Index: 1542742 (Python Code)

Max Activating Token Index: 593

Text #6

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.6719. Min Act: 0.0000

Data Index: 1174482 (C4 (Web Text))

Max Activating Token Index: 902

Text #7

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.6482. Min Act: -0.0000

Data Index: 191765 (C4 (Web Text))

Max Activating Token Index: 715

Text #8

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.7280. Min Act: -0.0000

Data Index: 1508119 (Python Code)

Max Activating Token Index: 689

Text #9

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.5673. Min Act: -0.0000

Data Index: 1567624 (Python Code)

Max Activating Token Index: 1008

Text #10

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.6745. Min Act: -0.0000

Data Index: 1629400 (Python Code)

Max Activating Token Index: 373

Text #11

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.6772. Min Act: -0.0000

Data Index: 1389885 (Python Code)

Max Activating Token Index: 614

Text #12

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.5352. Min Act: -0.0000

Data Index: 1635053 (Python Code)

Max Activating Token Index: 862

Text #13

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.5835. Min Act: -0.0000

Data Index: 588856 (C4 (Web Text))

Max Activating Token Index: 55

Text #14

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.4854. Min Act: 0.0000

Data Index: 998748 (C4 (Web Text))

Max Activating Token Index: 894

Text #15

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.3737. Min Act: -0.0000

Data Index: 1555766 (Python Code)

Max Activating Token Index: 865

Text #16

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.4315. Min Act: -0.0000

Data Index: 1558549 (Python Code)

Max Activating Token Index: 249

Text #17

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.4889. Min Act: -0.0000

Data Index: 1084877 (C4 (Web Text))

Max Activating Token Index: 283

Text #18

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.3909. Min Act: -0.0000

Data Index: 1492988 (Python Code)

Max Activating Token Index: 960

Text #19

Max Range: 2.3540. Min Range: -2.3540

Max Act: 1.3872. Min Act: -0.0000

Data Index: 1206492 (C4 (Web Text))

Max Activating Token Index: 381