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

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

Neuron 3578 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.5361. Min Act: -0.0000

Data Index: 1076389 (C4 (Web Text))

Max Activating Token Index: 166

Text #1

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.4369. Min Act: -0.0000

Data Index: 1468874 (Python Code)

Max Activating Token Index: 538

Text #2

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3995. Min Act: -0.0000

Data Index: 1620145 (Python Code)

Max Activating Token Index: 822

Text #3

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.4112. Min Act: -0.0000

Data Index: 1498913 (Python Code)

Max Activating Token Index: 633

Text #4

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3867. Min Act: -0.0000

Data Index: 596450 (C4 (Web Text))

Max Activating Token Index: 232

Text #5

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3918. Min Act: -0.0000

Data Index: 243986 (C4 (Web Text))

Max Activating Token Index: 14

Text #6

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3789. Min Act: -0.0000

Data Index: 697197 (C4 (Web Text))

Max Activating Token Index: 371

Text #7

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3845. Min Act: -0.0000

Data Index: 1163450 (C4 (Web Text))

Max Activating Token Index: 118

Text #8

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.4003. Min Act: -0.0000

Data Index: 678341 (C4 (Web Text))

Max Activating Token Index: 35

Text #9

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.4126. Min Act: -0.0000

Data Index: 1215186 (C4 (Web Text))

Max Activating Token Index: 136

Text #10

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3755. Min Act: -0.0000

Data Index: 96227 (C4 (Web Text))

Max Activating Token Index: 1014

Text #11

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3917. Min Act: -0.0000

Data Index: 1063591 (C4 (Web Text))

Max Activating Token Index: 25

Text #12

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3739. Min Act: -0.0000

Data Index: 1073112 (C4 (Web Text))

Max Activating Token Index: 150

Text #13

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3718. Min Act: -0.0000

Data Index: 1415994 (Python Code)

Max Activating Token Index: 616

Text #14

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3509. Min Act: -0.0000

Data Index: 257860 (C4 (Web Text))

Max Activating Token Index: 902

Text #15

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3380. Min Act: -0.0000

Data Index: 573861 (C4 (Web Text))

Max Activating Token Index: 87

Text #16

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3485. Min Act: -0.0000

Data Index: 1165147 (C4 (Web Text))

Max Activating Token Index: 151

Text #17

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3463. Min Act: -0.0000

Data Index: 190811 (C4 (Web Text))

Max Activating Token Index: 350

Text #18

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3823. Min Act: -0.0000

Data Index: 1184514 (C4 (Web Text))

Max Activating Token Index: 197

Text #19

Max Range: 0.5361. Min Range: -0.5361

Max Act: 0.3422. Min Act: -0.0000

Data Index: 347955 (C4 (Web Text))

Max Activating Token Index: 316