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

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

Neuron 3036 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: 0.7635. Min Range: -0.7635

Max Act: 0.7635. Min Act: -0.0000

Data Index: 645319 (C4 (Web Text))

Max Activating Token Index: 1005

Text #1

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.5602. Min Act: -0.0000

Data Index: 499270 (C4 (Web Text))

Max Activating Token Index: 442

Text #2

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.5469. Min Act: -0.0000

Data Index: 496169 (C4 (Web Text))

Max Activating Token Index: 255

Text #3

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.5687. Min Act: -0.0000

Data Index: 687383 (C4 (Web Text))

Max Activating Token Index: 588

Text #4

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.5459. Min Act: -0.0000

Data Index: 408390 (C4 (Web Text))

Max Activating Token Index: 436

Text #5

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.5212. Min Act: -0.0000

Data Index: 67735 (C4 (Web Text))

Max Activating Token Index: 456

Text #6

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.5209. Min Act: -0.0000

Data Index: 800018 (C4 (Web Text))

Max Activating Token Index: 559

Text #7

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4663. Min Act: -0.0000

Data Index: 647574 (C4 (Web Text))

Max Activating Token Index: 815

Text #8

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4652. Min Act: -0.0000

Data Index: 818357 (C4 (Web Text))

Max Activating Token Index: 660

Text #9

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4813. Min Act: -0.0000

Data Index: 792945 (C4 (Web Text))

Max Activating Token Index: 912

Text #10

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.3467. Min Act: -0.0000

Data Index: 167844 (C4 (Web Text))

Max Activating Token Index: 1010

Text #11

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.5083. Min Act: -0.0000

Data Index: 570753 (C4 (Web Text))

Max Activating Token Index: 452

Text #12

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4684. Min Act: -0.0000

Data Index: 1554855 (Python Code)

Max Activating Token Index: 15

Text #13

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4665. Min Act: -0.0000

Data Index: 1411459 (Python Code)

Max Activating Token Index: 62

Text #14

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4510. Min Act: -0.0000

Data Index: 1413225 (Python Code)

Max Activating Token Index: 6

Text #15

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4510. Min Act: -0.0000

Data Index: 1521039 (Python Code)

Max Activating Token Index: 6

Text #16

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4662. Min Act: -0.0000

Data Index: 1233270 (C4 (Web Text))

Max Activating Token Index: 69

Text #17

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4635. Min Act: -0.0000

Data Index: 673205 (C4 (Web Text))

Max Activating Token Index: 532

Text #18

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4608. Min Act: -0.0000

Data Index: 588980 (C4 (Web Text))

Max Activating Token Index: 1009

Text #19

Max Range: 0.7635. Min Range: -0.7635

Max Act: 0.4452. Min Act: -0.0000

Data Index: 307637 (C4 (Web Text))

Max Activating Token Index: 927