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

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

Neuron 5316 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: 0.9262. Min Range: -0.9262

Max Act: 0.9262. Min Act: -0.0000

Data Index: 1415193 (Python Code)

Max Activating Token Index: 4

Text #1

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.7707. Min Act: -0.0000

Data Index: 826527 (C4 (Web Text))

Max Activating Token Index: 353

Text #2

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.7104. Min Act: -0.0000

Data Index: 1002039 (C4 (Web Text))

Max Activating Token Index: 201

Text #3

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.7146. Min Act: -0.0000

Data Index: 546948 (C4 (Web Text))

Max Activating Token Index: 380

Text #4

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6047. Min Act: -0.0001

Data Index: 1548516 (Python Code)

Max Activating Token Index: 29

Text #5

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6419. Min Act: -0.0001

Data Index: 1539506 (Python Code)

Max Activating Token Index: 744

Text #6

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6790. Min Act: -0.0000

Data Index: 837684 (C4 (Web Text))

Max Activating Token Index: 542

Text #7

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6653. Min Act: -0.0000

Data Index: 1224254 (C4 (Web Text))

Max Activating Token Index: 185

Text #8

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6660. Min Act: -0.0000

Data Index: 1160427 (C4 (Web Text))

Max Activating Token Index: 229

Text #9

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6488. Min Act: -0.0000

Data Index: 726198 (C4 (Web Text))

Max Activating Token Index: 581

Text #10

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6396. Min Act: -0.0000

Data Index: 1574216 (Python Code)

Max Activating Token Index: 45

Text #11

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6384. Min Act: -0.0000

Data Index: 1509322 (Python Code)

Max Activating Token Index: 32

Text #12

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6125. Min Act: -0.0000

Data Index: 350893 (C4 (Web Text))

Max Activating Token Index: 935

Text #13

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.5810. Min Act: -0.0000

Data Index: 1637535 (Python Code)

Max Activating Token Index: 342

Text #14

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6111. Min Act: -0.0000

Data Index: 721938 (C4 (Web Text))

Max Activating Token Index: 524

Text #15

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.5654. Min Act: -0.0000

Data Index: 1581021 (Python Code)

Max Activating Token Index: 39

Text #16

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.6041. Min Act: -0.0000

Data Index: 253371 (C4 (Web Text))

Max Activating Token Index: 396

Text #17

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.5979. Min Act: -0.0000

Data Index: 1607012 (Python Code)

Max Activating Token Index: 24

Text #18

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.5745. Min Act: -0.0001

Data Index: 1420376 (Python Code)

Max Activating Token Index: 242

Text #19

Max Range: 0.9262. Min Range: -0.9262

Max Act: 0.5923. Min Act: -0.0000

Data Index: 624276 (C4 (Web Text))

Max Activating Token Index: 55