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

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

Neuron 3066 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 9.3996. Min Range: -9.3996

Max Act: 9.3267. Min Act: -0.0000

Data Index: 1571674 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 9.3996. Min Range: -9.3996

Max Act: 9.3996. Min Act: -0.0000

Data Index: 1447325 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 9.3996. Min Range: -9.3996

Max Act: 9.1437. Min Act: -0.0000

Data Index: 1393652 (Python Code)

Max Activating Token Index: 813

Text #3

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.7523. Min Act: -0.0000

Data Index: 26721 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 9.3996. Min Range: -9.3996

Max Act: 9.0789. Min Act: -0.0000

Data Index: 1566769 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.6062. Min Act: -0.0000

Data Index: 1504411 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.9028. Min Act: -0.0000

Data Index: 1566175 (Python Code)

Max Activating Token Index: 1023

Text #7

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.8678. Min Act: -0.0000

Data Index: 1531712 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.9688. Min Act: -0.0000

Data Index: 1383225 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.8228. Min Act: -0.0000

Data Index: 163433 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.9649. Min Act: -0.0000

Data Index: 1528032 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 9.3996. Min Range: -9.3996

Max Act: 9.0618. Min Act: -0.0000

Data Index: 1228931 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.5044. Min Act: -0.0000

Data Index: 1522966 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.3074. Min Act: -0.0000

Data Index: 1577552 (Python Code)

Max Activating Token Index: 1023

Text #14

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.2862. Min Act: -0.0000

Data Index: 1598943 (Python Code)

Max Activating Token Index: 1023

Text #15

Max Range: 9.3996. Min Range: -9.3996

Max Act: 7.9039. Min Act: -0.0000

Data Index: 1450787 (Python Code)

Max Activating Token Index: 1017

Text #16

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.3076. Min Act: -0.0000

Data Index: 1241902 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.1759. Min Act: -0.0000

Data Index: 615205 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 9.3996. Min Range: -9.3996

Max Act: 8.1263. Min Act: -0.0000

Data Index: 1652306 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 9.3996. Min Range: -9.3996

Max Act: 7.9091. Min Act: -0.0000

Data Index: 1412149 (Python Code)

Max Activating Token Index: 1023