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

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

Neuron 2782 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: 1.4888. Min Range: -1.4888

Max Act: 1.4639. Min Act: -0.0001

Data Index: 116398 (C4 (Web Text))

Max Activating Token Index: 150

Text #1

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.4888. Min Act: -0.0001

Data Index: 796579 (C4 (Web Text))

Max Activating Token Index: 758

Text #2

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.4551. Min Act: -0.0001

Data Index: 875189 (C4 (Web Text))

Max Activating Token Index: 827

Text #3

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.3690. Min Act: -0.0001

Data Index: 545394 (C4 (Web Text))

Max Activating Token Index: 354

Text #4

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.3967. Min Act: -0.0001

Data Index: 187408 (C4 (Web Text))

Max Activating Token Index: 400

Text #5

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.3769. Min Act: -0.0001

Data Index: 755938 (C4 (Web Text))

Max Activating Token Index: 846

Text #6

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.3393. Min Act: -0.0001

Data Index: 791104 (C4 (Web Text))

Max Activating Token Index: 930

Text #7

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.3327. Min Act: -0.0001

Data Index: 1466702 (Python Code)

Max Activating Token Index: 92

Text #8

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.3902. Min Act: -0.0001

Data Index: 223292 (C4 (Web Text))

Max Activating Token Index: 798

Text #9

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.3249. Min Act: -0.0001

Data Index: 1364690 (Python Code)

Max Activating Token Index: 101

Text #10

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.2557. Min Act: -0.0001

Data Index: 124348 (C4 (Web Text))

Max Activating Token Index: 710

Text #11

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.3354. Min Act: -0.0001

Data Index: 1476383 (Python Code)

Max Activating Token Index: 231

Text #12

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.2594. Min Act: -0.0001

Data Index: 1553067 (Python Code)

Max Activating Token Index: 508

Text #13

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.2901. Min Act: -0.0001

Data Index: 730051 (C4 (Web Text))

Max Activating Token Index: 309

Text #14

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.2901. Min Act: -0.0001

Data Index: 917564 (C4 (Web Text))

Max Activating Token Index: 751

Text #15

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.2445. Min Act: -0.0001

Data Index: 819457 (C4 (Web Text))

Max Activating Token Index: 55

Text #16

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.2820. Min Act: -0.0001

Data Index: 1613617 (Python Code)

Max Activating Token Index: 253

Text #17

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.2369. Min Act: -0.0001

Data Index: 1229251 (C4 (Web Text))

Max Activating Token Index: 120

Text #18

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.1881. Min Act: -0.0001

Data Index: 1253422 (C4 (Web Text))

Max Activating Token Index: 503

Text #19

Max Range: 1.4888. Min Range: -1.4888

Max Act: 1.2322. Min Act: -0.0001

Data Index: 905321 (C4 (Web Text))

Max Activating Token Index: 247