Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3072 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.7627. Min Act: -0.0001

Data Index: 184907 (C4 (Web Text))

Max Activating Token Index: 39

Text #1

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.7624. Min Act: -0.0001

Data Index: 1057131 (C4 (Web Text))

Max Activating Token Index: 50

Text #2

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.7568. Min Act: -0.0001

Data Index: 83635 (C4 (Web Text))

Max Activating Token Index: 48

Text #3

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.7653. Min Act: -0.0001

Data Index: 344083 (C4 (Web Text))

Max Activating Token Index: 45

Text #4

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.6278. Min Act: -0.0001

Data Index: 457901 (C4 (Web Text))

Max Activating Token Index: 41

Text #5

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.7533. Min Act: -0.0001

Data Index: 1240223 (C4 (Web Text))

Max Activating Token Index: 49

Text #6

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.3570. Min Act: -0.0001

Data Index: 958839 (C4 (Web Text))

Max Activating Token Index: 418

Text #7

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.3325. Min Act: -0.0001

Data Index: 390605 (C4 (Web Text))

Max Activating Token Index: 48

Text #8

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.3397. Min Act: -0.0001

Data Index: 1343705 (C4 (Web Text))

Max Activating Token Index: 45

Text #9

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.3470. Min Act: -0.0001

Data Index: 1200507 (C4 (Web Text))

Max Activating Token Index: 382

Text #10

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.4869. Min Act: -0.0001

Data Index: 999457 (C4 (Web Text))

Max Activating Token Index: 33

Text #11

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.4420. Min Act: -0.0001

Data Index: 1348712 (C4 (Web Text))

Max Activating Token Index: 74

Text #12

Max Range: 3.7653. Min Range: -3.7653

Max Act: 2.9284. Min Act: -0.0001

Data Index: 1132908 (C4 (Web Text))

Max Activating Token Index: 35

Text #13

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.0547. Min Act: -0.0001

Data Index: 1164220 (C4 (Web Text))

Max Activating Token Index: 470

Text #14

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.1844. Min Act: -0.0001

Data Index: 742178 (C4 (Web Text))

Max Activating Token Index: 52

Text #15

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.2529. Min Act: -0.0001

Data Index: 282643 (C4 (Web Text))

Max Activating Token Index: 282

Text #16

Max Range: 3.7653. Min Range: -3.7653

Max Act: 2.9075. Min Act: -0.0001

Data Index: 239117 (C4 (Web Text))

Max Activating Token Index: 34

Text #17

Max Range: 3.7653. Min Range: -3.7653

Max Act: 3.0519. Min Act: -0.0001

Data Index: 401854 (C4 (Web Text))

Max Activating Token Index: 31

Text #18

Max Range: 3.7653. Min Range: -3.7653

Max Act: 2.8876. Min Act: -0.0001

Data Index: 835918 (C4 (Web Text))

Max Activating Token Index: 31

Text #19

Max Range: 3.7653. Min Range: -3.7653

Max Act: 2.8772. Min Act: -0.0001

Data Index: 605119 (C4 (Web Text))

Max Activating Token Index: 18