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

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

Neuron 4812 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1140. Min Range: -3.1140

Max Act: 3.1140. Min Act: -0.0001

Data Index: 1532589 (Python Code)

Max Activating Token Index: 814

Text #1

Max Range: 3.1140. Min Range: -3.1140

Max Act: 2.8725. Min Act: -0.0001

Data Index: 1402613 (Python Code)

Max Activating Token Index: 298

Text #2

Max Range: 3.1140. Min Range: -3.1140

Max Act: 2.3339. Min Act: -0.0001

Data Index: 1615423 (Python Code)

Max Activating Token Index: 406

Text #3

Max Range: 3.1140. Min Range: -3.1140

Max Act: 2.3629. Min Act: -0.0001

Data Index: 1544240 (Python Code)

Max Activating Token Index: 374

Text #4

Max Range: 3.1140. Min Range: -3.1140

Max Act: 2.1490. Min Act: -0.0001

Data Index: 1567555 (Python Code)

Max Activating Token Index: 168

Text #5

Max Range: 3.1140. Min Range: -3.1140

Max Act: 2.1144. Min Act: -0.0001

Data Index: 1624390 (Python Code)

Max Activating Token Index: 935

Text #6

Max Range: 3.1140. Min Range: -3.1140

Max Act: 2.1816. Min Act: -0.0001

Data Index: 1625899 (Python Code)

Max Activating Token Index: 789

Text #7

Max Range: 3.1140. Min Range: -3.1140

Max Act: 2.1065. Min Act: -0.0001

Data Index: 1462546 (Python Code)

Max Activating Token Index: 745

Text #8

Max Range: 3.1140. Min Range: -3.1140

Max Act: 2.0030. Min Act: -0.0001

Data Index: 1559285 (Python Code)

Max Activating Token Index: 337

Text #9

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.9457. Min Act: -0.0001

Data Index: 1530237 (Python Code)

Max Activating Token Index: 525

Text #10

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.9258. Min Act: -0.0001

Data Index: 1501759 (Python Code)

Max Activating Token Index: 449

Text #11

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.9744. Min Act: -0.0001

Data Index: 1416865 (Python Code)

Max Activating Token Index: 905

Text #12

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.8231. Min Act: -0.0001

Data Index: 1622309 (Python Code)

Max Activating Token Index: 1000

Text #13

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.9783. Min Act: -0.0001

Data Index: 1382675 (Python Code)

Max Activating Token Index: 149

Text #14

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.8730. Min Act: -0.0001

Data Index: 1587912 (Python Code)

Max Activating Token Index: 456

Text #15

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.8394. Min Act: -0.0001

Data Index: 1644554 (Python Code)

Max Activating Token Index: 241

Text #16

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.6064. Min Act: -0.0000

Data Index: 1493487 (Python Code)

Max Activating Token Index: 900

Text #17

Max Range: 3.1140. Min Range: -3.1140

Max Act: 2.0430. Min Act: -0.0001

Data Index: 1619060 (Python Code)

Max Activating Token Index: 930

Text #18

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.7870. Min Act: -0.0001

Data Index: 1312230 (C4 (Web Text))

Max Activating Token Index: 972

Text #19

Max Range: 3.1140. Min Range: -3.1140

Max Act: 1.6793. Min Act: -0.0001

Data Index: 1551775 (Python Code)

Max Activating Token Index: 240