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

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

Neuron 4348 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.8119. Min Act: -0.0000

Data Index: 512678 (C4 (Web Text))

Max Activating Token Index: 486

Text #1

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.4137. Min Act: -0.0000

Data Index: 631042 (C4 (Web Text))

Max Activating Token Index: 221

Text #2

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.5260. Min Act: -0.0000

Data Index: 1589121 (Python Code)

Max Activating Token Index: 476

Text #3

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.5268. Min Act: -0.0000

Data Index: 1371201 (Python Code)

Max Activating Token Index: 1003

Text #4

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.4729. Min Act: -0.0000

Data Index: 1070390 (C4 (Web Text))

Max Activating Token Index: 307

Text #5

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.4653. Min Act: -0.0001

Data Index: 1394856 (Python Code)

Max Activating Token Index: 646

Text #6

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.3365. Min Act: -0.0000

Data Index: 646488 (C4 (Web Text))

Max Activating Token Index: 461

Text #7

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.1984. Min Act: -0.0000

Data Index: 1555394 (Python Code)

Max Activating Token Index: 146

Text #8

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.3098. Min Act: -0.0000

Data Index: 688662 (C4 (Web Text))

Max Activating Token Index: 570

Text #9

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.2743. Min Act: -0.0000

Data Index: 1478137 (Python Code)

Max Activating Token Index: 510

Text #10

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.3555. Min Act: -0.0000

Data Index: 1049032 (C4 (Web Text))

Max Activating Token Index: 547

Text #11

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.2543. Min Act: -0.0000

Data Index: 1360231 (Python Code)

Max Activating Token Index: 452

Text #12

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.2465. Min Act: -0.0000

Data Index: 1520140 (Python Code)

Max Activating Token Index: 328

Text #13

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.2073. Min Act: -0.0001

Data Index: 1523718 (Python Code)

Max Activating Token Index: 110

Text #14

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.0870. Min Act: -0.0000

Data Index: 448953 (C4 (Web Text))

Max Activating Token Index: 416

Text #15

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.1735. Min Act: -0.0000

Data Index: 337998 (C4 (Web Text))

Max Activating Token Index: 237

Text #16

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.3112. Min Act: -0.0000

Data Index: 434316 (C4 (Web Text))

Max Activating Token Index: 65

Text #17

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.0210. Min Act: -0.0001

Data Index: 1369461 (Python Code)

Max Activating Token Index: 936

Text #18

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.0277. Min Act: -0.0000

Data Index: 1585493 (Python Code)

Max Activating Token Index: 784

Text #19

Max Range: 2.8119. Min Range: -2.8119

Max Act: 2.3345. Min Act: -0.0001

Data Index: 1473162 (Python Code)

Max Activating Token Index: 781