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

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

Neuron 3158 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.7145. Min Act: -0.0002

Data Index: 1459167 (Python Code)

Max Activating Token Index: 81

Text #1

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.7078. Min Act: -0.0002

Data Index: 1560324 (Python Code)

Max Activating Token Index: 736

Text #2

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.7073. Min Act: -0.0002

Data Index: 1360618 (Python Code)

Max Activating Token Index: 9

Text #3

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.7073. Min Act: -0.0002

Data Index: 1388801 (Python Code)

Max Activating Token Index: 9

Text #4

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.7075. Min Act: -0.0002

Data Index: 1454918 (Python Code)

Max Activating Token Index: 24

Text #5

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.7043. Min Act: -0.0002

Data Index: 1474966 (Python Code)

Max Activating Token Index: 7

Text #6

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6734. Min Act: -0.0002

Data Index: 1422584 (Python Code)

Max Activating Token Index: 3

Text #7

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6472. Min Act: -0.0002

Data Index: 1551800 (Python Code)

Max Activating Token Index: 50

Text #8

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6888. Min Act: -0.0002

Data Index: 1379148 (Python Code)

Max Activating Token Index: 216

Text #9

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6597. Min Act: -0.0002

Data Index: 1560390 (Python Code)

Max Activating Token Index: 311

Text #10

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6797. Min Act: -0.0002

Data Index: 1613732 (Python Code)

Max Activating Token Index: 28

Text #11

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6573. Min Act: -0.0002

Data Index: 1460502 (Python Code)

Max Activating Token Index: 290

Text #12

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.5783. Min Act: -0.0002

Data Index: 1382323 (Python Code)

Max Activating Token Index: 198

Text #13

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6730. Min Act: -0.0002

Data Index: 1525485 (Python Code)

Max Activating Token Index: 29

Text #14

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6916. Min Act: -0.0002

Data Index: 1506205 (Python Code)

Max Activating Token Index: 404

Text #15

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6277. Min Act: -0.0002

Data Index: 1608411 (Python Code)

Max Activating Token Index: 237

Text #16

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6646. Min Act: -0.0002

Data Index: 1559536 (Python Code)

Max Activating Token Index: 959

Text #17

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6406. Min Act: -0.0002

Data Index: 1521704 (Python Code)

Max Activating Token Index: 429

Text #18

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6671. Min Act: -0.0002

Data Index: 1481966 (Python Code)

Max Activating Token Index: 123

Text #19

Max Range: 0.7145. Min Range: -0.7145

Max Act: 0.6374. Min Act: -0.0002

Data Index: 1475808 (Python Code)

Max Activating Token Index: 783