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

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

Neuron 4472 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.6626. Min Act: -0.0000

Data Index: 1576999 (Python Code)

Max Activating Token Index: 654

Text #1

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.6397. Min Act: -0.0000

Data Index: 1624983 (Python Code)

Max Activating Token Index: 523

Text #2

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.6102. Min Act: -0.0000

Data Index: 1344480 (C4 (Web Text))

Max Activating Token Index: 449

Text #3

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.5836. Min Act: -0.0000

Data Index: 1346584 (C4 (Web Text))

Max Activating Token Index: 720

Text #4

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.5552. Min Act: -0.0000

Data Index: 1631715 (Python Code)

Max Activating Token Index: 236

Text #5

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.2910. Min Act: -0.0000

Data Index: 110153 (C4 (Web Text))

Max Activating Token Index: 762

Text #6

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.3541. Min Act: -0.0000

Data Index: 1485030 (Python Code)

Max Activating Token Index: 737

Text #7

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.3441. Min Act: -0.0000

Data Index: 323142 (C4 (Web Text))

Max Activating Token Index: 765

Text #8

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.4081. Min Act: -0.0000

Data Index: 1509483 (Python Code)

Max Activating Token Index: 1018

Text #9

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.3151. Min Act: -0.0000

Data Index: 47858 (C4 (Web Text))

Max Activating Token Index: 113

Text #10

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.3027. Min Act: -0.0000

Data Index: 341151 (C4 (Web Text))

Max Activating Token Index: 54

Text #11

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.2554. Min Act: -0.0000

Data Index: 785108 (C4 (Web Text))

Max Activating Token Index: 516

Text #12

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.3161. Min Act: -0.0000

Data Index: 1486849 (Python Code)

Max Activating Token Index: 225

Text #13

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.2135. Min Act: -0.0000

Data Index: 1293027 (C4 (Web Text))

Max Activating Token Index: 361

Text #14

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.1435. Min Act: -0.0000

Data Index: 1184094 (C4 (Web Text))

Max Activating Token Index: 645

Text #15

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.1949. Min Act: -0.0000

Data Index: 1190152 (C4 (Web Text))

Max Activating Token Index: 198

Text #16

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.1132. Min Act: -0.0000

Data Index: 1548897 (Python Code)

Max Activating Token Index: 741

Text #17

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.1742. Min Act: -0.0000

Data Index: 490412 (C4 (Web Text))

Max Activating Token Index: 584

Text #18

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.2425. Min Act: -0.0000

Data Index: 1490855 (Python Code)

Max Activating Token Index: 447

Text #19

Max Range: 1.6626. Min Range: -1.6626

Max Act: 1.2242. Min Act: -0.0000

Data Index: 1278303 (C4 (Web Text))

Max Activating Token Index: 789