Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2930 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.9534. Min Act: -0.0000

Data Index: 1158757 (C4 (Web Text))

Max Activating Token Index: 518

Text #1

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.8390. Min Act: -0.0000

Data Index: 427571 (C4 (Web Text))

Max Activating Token Index: 680

Text #2

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.7489. Min Act: -0.0000

Data Index: 171275 (C4 (Web Text))

Max Activating Token Index: 347

Text #3

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5808. Min Act: -0.0000

Data Index: 1481793 (Python Code)

Max Activating Token Index: 765

Text #4

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.6139. Min Act: -0.0000

Data Index: 1110856 (C4 (Web Text))

Max Activating Token Index: 418

Text #5

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.6176. Min Act: -0.0000

Data Index: 1600959 (Python Code)

Max Activating Token Index: 143

Text #6

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5772. Min Act: -0.0000

Data Index: 1377454 (Python Code)

Max Activating Token Index: 942

Text #7

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.6002. Min Act: -0.0000

Data Index: 1393952 (Python Code)

Max Activating Token Index: 865

Text #8

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.6064. Min Act: -0.0000

Data Index: 1459075 (Python Code)

Max Activating Token Index: 676

Text #9

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.6060. Min Act: -0.0000

Data Index: 1256553 (C4 (Web Text))

Max Activating Token Index: 300

Text #10

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5495. Min Act: -0.0000

Data Index: 1467611 (Python Code)

Max Activating Token Index: 228

Text #11

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5145. Min Act: -0.0000

Data Index: 1422534 (Python Code)

Max Activating Token Index: 40

Text #12

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5148. Min Act: -0.0000

Data Index: 1323310 (C4 (Web Text))

Max Activating Token Index: 339

Text #13

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5395. Min Act: -0.0000

Data Index: 1406899 (Python Code)

Max Activating Token Index: 954

Text #14

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5340. Min Act: -0.0000

Data Index: 1587494 (Python Code)

Max Activating Token Index: 399

Text #15

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.4668. Min Act: -0.0000

Data Index: 1365396 (Python Code)

Max Activating Token Index: 794

Text #16

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5336. Min Act: -0.0000

Data Index: 22592 (C4 (Web Text))

Max Activating Token Index: 602

Text #17

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5077. Min Act: -0.0000

Data Index: 1185805 (C4 (Web Text))

Max Activating Token Index: 357

Text #18

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5151. Min Act: -0.0000

Data Index: 1183034 (C4 (Web Text))

Max Activating Token Index: 174

Text #19

Max Range: 0.9534. Min Range: -0.9534

Max Act: 0.5303. Min Act: -0.0000

Data Index: 1242033 (C4 (Web Text))

Max Activating Token Index: 388