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

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

Neuron 2049 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1542. Min Range: -5.1542

Max Act: 4.7868. Min Act: -0.0000

Data Index: 1178534 (C4 (Web Text))

Max Activating Token Index: 963

Text #1

Max Range: 5.1542. Min Range: -5.1542

Max Act: 5.1542. Min Act: -0.0000

Data Index: 945807 (C4 (Web Text))

Max Activating Token Index: 548

Text #2

Max Range: 5.1542. Min Range: -5.1542

Max Act: 4.6285. Min Act: -0.0000

Data Index: 948664 (C4 (Web Text))

Max Activating Token Index: 698

Text #3

Max Range: 5.1542. Min Range: -5.1542

Max Act: 4.0503. Min Act: -0.0000

Data Index: 630220 (C4 (Web Text))

Max Activating Token Index: 427

Text #4

Max Range: 5.1542. Min Range: -5.1542

Max Act: 4.2121. Min Act: -0.0000

Data Index: 1540051 (Python Code)

Max Activating Token Index: 874

Text #5

Max Range: 5.1542. Min Range: -5.1542

Max Act: 3.7658. Min Act: -0.0000

Data Index: 550777 (C4 (Web Text))

Max Activating Token Index: 615

Text #6

Max Range: 5.1542. Min Range: -5.1542

Max Act: 3.8761. Min Act: -0.0000

Data Index: 75048 (C4 (Web Text))

Max Activating Token Index: 375

Text #7

Max Range: 5.1542. Min Range: -5.1542

Max Act: 3.7405. Min Act: -0.0000

Data Index: 25256 (C4 (Web Text))

Max Activating Token Index: 398

Text #8

Max Range: 5.1542. Min Range: -5.1542

Max Act: 3.6774. Min Act: -0.0000

Data Index: 1124476 (C4 (Web Text))

Max Activating Token Index: 334

Text #9

Max Range: 5.1542. Min Range: -5.1542

Max Act: 3.6452. Min Act: -0.0000

Data Index: 870677 (C4 (Web Text))

Max Activating Token Index: 178

Text #10

Max Range: 5.1542. Min Range: -5.1542

Max Act: 3.3898. Min Act: -0.0000

Data Index: 917967 (C4 (Web Text))

Max Activating Token Index: 756

Text #11

Max Range: 5.1542. Min Range: -5.1542

Max Act: 3.4497. Min Act: -0.0000

Data Index: 582008 (C4 (Web Text))

Max Activating Token Index: 623

Text #12

Max Range: 5.1542. Min Range: -5.1542

Max Act: 3.1359. Min Act: -0.0000

Data Index: 1494872 (Python Code)

Max Activating Token Index: 82

Text #13

Max Range: 5.1542. Min Range: -5.1542

Max Act: 3.0015. Min Act: -0.0000

Data Index: 1280963 (C4 (Web Text))

Max Activating Token Index: 930

Text #14

Max Range: 5.1542. Min Range: -5.1542

Max Act: 2.7876. Min Act: -0.0000

Data Index: 86333 (C4 (Web Text))

Max Activating Token Index: 432

Text #15

Max Range: 5.1542. Min Range: -5.1542

Max Act: 2.9145. Min Act: -0.0000

Data Index: 921339 (C4 (Web Text))

Max Activating Token Index: 980

Text #16

Max Range: 5.1542. Min Range: -5.1542

Max Act: 2.9954. Min Act: -0.0000

Data Index: 974577 (C4 (Web Text))

Max Activating Token Index: 133

Text #17

Max Range: 5.1542. Min Range: -5.1542

Max Act: 2.6508. Min Act: -0.0000

Data Index: 809150 (C4 (Web Text))

Max Activating Token Index: 762

Text #18

Max Range: 5.1542. Min Range: -5.1542

Max Act: 2.7134. Min Act: -0.0000

Data Index: 715059 (C4 (Web Text))

Max Activating Token Index: 68

Text #19

Max Range: 5.1542. Min Range: -5.1542

Max Act: 2.6904. Min Act: -0.0000

Data Index: 1613248 (Python Code)

Max Activating Token Index: 372