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

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

Neuron 4649 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.9502. Min Act: -0.0001

Data Index: 1593758 (Python Code)

Max Activating Token Index: 6

Text #1

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.9502. Min Act: -0.0001

Data Index: 1447154 (Python Code)

Max Activating Token Index: 6

Text #2

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.9711. Min Act: -0.0001

Data Index: 1439730 (Python Code)

Max Activating Token Index: 10

Text #3

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.9001. Min Act: -0.0001

Data Index: 1362177 (Python Code)

Max Activating Token Index: 10

Text #4

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.9781. Min Act: -0.0001

Data Index: 1492168 (Python Code)

Max Activating Token Index: 12

Text #5

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.8814. Min Act: -0.0001

Data Index: 1472461 (Python Code)

Max Activating Token Index: 14

Text #6

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.7061. Min Act: -0.0001

Data Index: 1478663 (Python Code)

Max Activating Token Index: 31

Text #7

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.8174. Min Act: -0.0001

Data Index: 1507426 (Python Code)

Max Activating Token Index: 60

Text #8

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.7845. Min Act: -0.0001

Data Index: 1475970 (Python Code)

Max Activating Token Index: 12

Text #9

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.6851. Min Act: -0.0001

Data Index: 1427603 (Python Code)

Max Activating Token Index: 18

Text #10

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.6776. Min Act: -0.0001

Data Index: 1650908 (Python Code)

Max Activating Token Index: 38

Text #11

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.6883. Min Act: -0.0001

Data Index: 1502365 (Python Code)

Max Activating Token Index: 13

Text #12

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.7029. Min Act: -0.0001

Data Index: 1648358 (Python Code)

Max Activating Token Index: 9

Text #13

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.6021. Min Act: -0.0001

Data Index: 1435328 (Python Code)

Max Activating Token Index: 16

Text #14

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.6100. Min Act: -0.0001

Data Index: 1372873 (Python Code)

Max Activating Token Index: 11

Text #15

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.5277. Min Act: -0.0002

Data Index: 1477058 (Python Code)

Max Activating Token Index: 52

Text #16

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.4881. Min Act: -0.0001

Data Index: 1625342 (Python Code)

Max Activating Token Index: 19

Text #17

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.5268. Min Act: -0.0001

Data Index: 1565325 (Python Code)

Max Activating Token Index: 30

Text #18

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.4676. Min Act: -0.0001

Data Index: 1576745 (Python Code)

Max Activating Token Index: 14

Text #19

Max Range: 6.9781. Min Range: -6.9781

Max Act: 6.5658. Min Act: -0.0001

Data Index: 1600325 (Python Code)

Max Activating Token Index: 13