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

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

Neuron 5659 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.4845. Min Act: -0.0000

Data Index: 222249 (C4 (Web Text))

Max Activating Token Index: 154

Text #1

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.4770. Min Act: -0.0000

Data Index: 994149 (C4 (Web Text))

Max Activating Token Index: 459

Text #2

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.4483. Min Act: -0.0000

Data Index: 1161373 (C4 (Web Text))

Max Activating Token Index: 42

Text #3

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.4141. Min Act: -0.0001

Data Index: 1499641 (Python Code)

Max Activating Token Index: 327

Text #4

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.4023. Min Act: -0.0000

Data Index: 910870 (C4 (Web Text))

Max Activating Token Index: 354

Text #5

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.4111. Min Act: -0.0000

Data Index: 319964 (C4 (Web Text))

Max Activating Token Index: 990

Text #6

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3987. Min Act: -0.0000

Data Index: 527480 (C4 (Web Text))

Max Activating Token Index: 648

Text #7

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3691. Min Act: -0.0000

Data Index: 1104465 (C4 (Web Text))

Max Activating Token Index: 683

Text #8

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3910. Min Act: -0.0000

Data Index: 1238557 (C4 (Web Text))

Max Activating Token Index: 537

Text #9

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3878. Min Act: -0.0001

Data Index: 1168273 (C4 (Web Text))

Max Activating Token Index: 739

Text #10

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3529. Min Act: -0.0000

Data Index: 830326 (C4 (Web Text))

Max Activating Token Index: 710

Text #11

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3660. Min Act: -0.0000

Data Index: 1278339 (C4 (Web Text))

Max Activating Token Index: 148

Text #12

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3607. Min Act: -0.0000

Data Index: 780370 (C4 (Web Text))

Max Activating Token Index: 282

Text #13

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3611. Min Act: -0.0000

Data Index: 1217659 (C4 (Web Text))

Max Activating Token Index: 611

Text #14

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3648. Min Act: -0.0001

Data Index: 384712 (C4 (Web Text))

Max Activating Token Index: 355

Text #15

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3495. Min Act: -0.0000

Data Index: 1581716 (Python Code)

Max Activating Token Index: 317

Text #16

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3455. Min Act: -0.0000

Data Index: 254592 (C4 (Web Text))

Max Activating Token Index: 461

Text #17

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3309. Min Act: -0.0000

Data Index: 32024 (C4 (Web Text))

Max Activating Token Index: 227

Text #18

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3321. Min Act: -0.0000

Data Index: 471255 (C4 (Web Text))

Max Activating Token Index: 999

Text #19

Max Range: 0.4845. Min Range: -0.4845

Max Act: 0.3240. Min Act: -0.0000

Data Index: 1168413 (C4 (Web Text))

Max Activating Token Index: 98