Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 476 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.7496. Min Act: -0.0006

Data Index: 778393 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.7042. Min Act: -0.0005

Data Index: 1547847 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.7004. Min Act: -0.0006

Data Index: 1100306 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6878. Min Act: -0.0006

Data Index: 1418219 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6687. Min Act: -0.0005

Data Index: 1546129 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6598. Min Act: -0.0006

Data Index: 1397516 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6578. Min Act: -0.0006

Data Index: 771679 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6579. Min Act: -0.0006

Data Index: 96672 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6478. Min Act: -0.0006

Data Index: 1406928 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6442. Min Act: -0.0006

Data Index: 1519894 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6389. Min Act: -0.0006

Data Index: 1462039 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6379. Min Act: -0.0006

Data Index: 429208 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6576. Min Act: -0.0006

Data Index: 1644956 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6360. Min Act: -0.0006

Data Index: 1496701 (Python Code)

Max Activating Token Index: 1023

Text #14

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6323. Min Act: -0.0006

Data Index: 1181803 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6301. Min Act: -0.0006

Data Index: 1369600 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6311. Min Act: -0.0005

Data Index: 1506290 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6285. Min Act: -0.0006

Data Index: 1132322 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6254. Min Act: -0.0005

Data Index: 888259 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6435. Min Act: -0.0005

Data Index: 1374528 (Python Code)

Max Activating Token Index: 9