Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3447 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.7227. Min Act: 0.0000

Data Index: 1584059 (Python Code)

Max Activating Token Index: 284

Text #1

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6973. Min Act: -0.0000

Data Index: 1492813 (Python Code)

Max Activating Token Index: 94

Text #2

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.7531. Min Act: 0.0000

Data Index: 1414610 (Python Code)

Max Activating Token Index: 306

Text #3

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6946. Min Act: -0.0000

Data Index: 1368700 (Python Code)

Max Activating Token Index: 47

Text #4

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6589. Min Act: 0.0000

Data Index: 1641305 (Python Code)

Max Activating Token Index: 213

Text #5

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6454. Min Act: -0.0001

Data Index: 1441077 (Python Code)

Max Activating Token Index: 584

Text #6

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6436. Min Act: -0.0000

Data Index: 1482361 (Python Code)

Max Activating Token Index: 171

Text #7

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6139. Min Act: 0.0001

Data Index: 1620265 (Python Code)

Max Activating Token Index: 463

Text #8

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6304. Min Act: 0.0000

Data Index: 1635464 (Python Code)

Max Activating Token Index: 67

Text #9

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6082. Min Act: 0.0000

Data Index: 1655515 (Python Code)

Max Activating Token Index: 80

Text #10

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6057. Min Act: 0.0000

Data Index: 1562222 (Python Code)

Max Activating Token Index: 29

Text #11

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6276. Min Act: 0.0000

Data Index: 1386041 (Python Code)

Max Activating Token Index: 320

Text #12

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.6052. Min Act: -0.0000

Data Index: 1608651 (Python Code)

Max Activating Token Index: 433

Text #13

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.5760. Min Act: -0.0001

Data Index: 1565533 (Python Code)

Max Activating Token Index: 66

Text #14

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.5725. Min Act: -0.0001

Data Index: 1394666 (Python Code)

Max Activating Token Index: 65

Text #15

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.5813. Min Act: -0.0000

Data Index: 1609519 (Python Code)

Max Activating Token Index: 445

Text #16

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.5649. Min Act: -0.0001

Data Index: 1633281 (Python Code)

Max Activating Token Index: 67

Text #17

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.5817. Min Act: 0.0000

Data Index: 1627358 (Python Code)

Max Activating Token Index: 921

Text #18

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.5624. Min Act: -0.0000

Data Index: 1656894 (Python Code)

Max Activating Token Index: 923

Text #19

Max Range: 0.7531. Min Range: -0.7531

Max Act: 0.5410. Min Act: -0.0000

Data Index: 1459110 (Python Code)

Max Activating Token Index: 717