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

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

Neuron 4640 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9837. Min Range: -3.9837

Max Act: 3.9837. Min Act: -0.0000

Data Index: 92570 (C4 (Web Text))

Max Activating Token Index: 16

Text #1

Max Range: 3.9837. Min Range: -3.9837

Max Act: 2.9235. Min Act: -0.0000

Data Index: 504336 (C4 (Web Text))

Max Activating Token Index: 334

Text #2

Max Range: 3.9837. Min Range: -3.9837

Max Act: 2.8088. Min Act: -0.0000

Data Index: 101618 (C4 (Web Text))

Max Activating Token Index: 104

Text #3

Max Range: 3.9837. Min Range: -3.9837

Max Act: 2.0932. Min Act: -0.0000

Data Index: 252499 (C4 (Web Text))

Max Activating Token Index: 117

Text #4

Max Range: 3.9837. Min Range: -3.9837

Max Act: 2.4340. Min Act: -0.0000

Data Index: 293765 (C4 (Web Text))

Max Activating Token Index: 80

Text #5

Max Range: 3.9837. Min Range: -3.9837

Max Act: 2.3929. Min Act: -0.0000

Data Index: 926336 (C4 (Web Text))

Max Activating Token Index: 170

Text #6

Max Range: 3.9837. Min Range: -3.9837

Max Act: 2.4944. Min Act: -0.0000

Data Index: 1227404 (C4 (Web Text))

Max Activating Token Index: 136

Text #7

Max Range: 3.9837. Min Range: -3.9837

Max Act: 2.1496. Min Act: -0.0000

Data Index: 480301 (C4 (Web Text))

Max Activating Token Index: 85

Text #8

Max Range: 3.9837. Min Range: -3.9837

Max Act: 2.2233. Min Act: -0.0000

Data Index: 688828 (C4 (Web Text))

Max Activating Token Index: 242

Text #9

Max Range: 3.9837. Min Range: -3.9837

Max Act: 1.9547. Min Act: -0.0000

Data Index: 308855 (C4 (Web Text))

Max Activating Token Index: 546

Text #10

Max Range: 3.9837. Min Range: -3.9837

Max Act: 1.9403. Min Act: -0.0000

Data Index: 776639 (C4 (Web Text))

Max Activating Token Index: 725

Text #11

Max Range: 3.9837. Min Range: -3.9837

Max Act: 1.9501. Min Act: -0.0000

Data Index: 1149034 (C4 (Web Text))

Max Activating Token Index: 632

Text #12

Max Range: 3.9837. Min Range: -3.9837

Max Act: 1.7661. Min Act: -0.0000

Data Index: 376235 (C4 (Web Text))

Max Activating Token Index: 279

Text #13

Max Range: 3.9837. Min Range: -3.9837

Max Act: 1.7158. Min Act: -0.0000

Data Index: 315404 (C4 (Web Text))

Max Activating Token Index: 115

Text #14

Max Range: 3.9837. Min Range: -3.9837

Max Act: 2.1870. Min Act: -0.0000

Data Index: 712879 (C4 (Web Text))

Max Activating Token Index: 415

Text #15

Max Range: 3.9837. Min Range: -3.9837

Max Act: 1.6722. Min Act: -0.0000

Data Index: 951967 (C4 (Web Text))

Max Activating Token Index: 726

Text #16

Max Range: 3.9837. Min Range: -3.9837

Max Act: 1.1406. Min Act: -0.0000

Data Index: 1421414 (Python Code)

Max Activating Token Index: 527

Text #17

Max Range: 3.9837. Min Range: -3.9837

Max Act: 1.0281. Min Act: -0.0000

Data Index: 176153 (C4 (Web Text))

Max Activating Token Index: 88

Text #18

Max Range: 3.9837. Min Range: -3.9837

Max Act: 0.9645. Min Act: -0.0000

Data Index: 1575182 (Python Code)

Max Activating Token Index: 99

Text #19

Max Range: 3.9837. Min Range: -3.9837

Max Act: 1.0016. Min Act: -0.0000

Data Index: 262594 (C4 (Web Text))

Max Activating Token Index: 280