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

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

Neuron 2285 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.9948. Min Act: -0.0000

Data Index: 1617127 (Python Code)

Max Activating Token Index: 141

Text #1

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.9265. Min Act: -0.0001

Data Index: 994792 (C4 (Web Text))

Max Activating Token Index: 37

Text #2

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8141. Min Act: -0.0001

Data Index: 232243 (C4 (Web Text))

Max Activating Token Index: 242

Text #3

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.9470. Min Act: -0.0000

Data Index: 1555877 (Python Code)

Max Activating Token Index: 946

Text #4

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.9396. Min Act: -0.0001

Data Index: 13694 (C4 (Web Text))

Max Activating Token Index: 466

Text #5

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8801. Min Act: -0.0001

Data Index: 317483 (C4 (Web Text))

Max Activating Token Index: 276

Text #6

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8899. Min Act: -0.0001

Data Index: 1304603 (C4 (Web Text))

Max Activating Token Index: 213

Text #7

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8380. Min Act: -0.0001

Data Index: 535765 (C4 (Web Text))

Max Activating Token Index: 936

Text #8

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8380. Min Act: -0.0001

Data Index: 1230616 (C4 (Web Text))

Max Activating Token Index: 162

Text #9

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8597. Min Act: -0.0001

Data Index: 1502599 (Python Code)

Max Activating Token Index: 263

Text #10

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8888. Min Act: -0.0000

Data Index: 1504765 (Python Code)

Max Activating Token Index: 916

Text #11

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8495. Min Act: -0.0001

Data Index: 1523333 (Python Code)

Max Activating Token Index: 398

Text #12

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8896. Min Act: -0.0001

Data Index: 1260493 (C4 (Web Text))

Max Activating Token Index: 902

Text #13

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8049. Min Act: -0.0001

Data Index: 1206894 (C4 (Web Text))

Max Activating Token Index: 710

Text #14

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8260. Min Act: -0.0001

Data Index: 1033103 (C4 (Web Text))

Max Activating Token Index: 898

Text #15

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.9745. Min Act: -0.0001

Data Index: 349572 (C4 (Web Text))

Max Activating Token Index: 781

Text #16

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8141. Min Act: -0.0001

Data Index: 55099 (C4 (Web Text))

Max Activating Token Index: 296

Text #17

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8524. Min Act: -0.0001

Data Index: 1567667 (Python Code)

Max Activating Token Index: 501

Text #18

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8727. Min Act: -0.0000

Data Index: 1545996 (Python Code)

Max Activating Token Index: 225

Text #19

Max Range: 0.9948. Min Range: -0.9948

Max Act: 0.8289. Min Act: -0.0001

Data Index: 605800 (C4 (Web Text))

Max Activating Token Index: 560