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

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

Neuron 4105 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.5722. Min Act: -0.0000

Data Index: 863558 (C4 (Web Text))

Max Activating Token Index: 80

Text #1

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.3424. Min Act: -0.0000

Data Index: 1304618 (C4 (Web Text))

Max Activating Token Index: 229

Text #2

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.4377. Min Act: -0.0000

Data Index: 678755 (C4 (Web Text))

Max Activating Token Index: 315

Text #3

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.3152. Min Act: -0.0000

Data Index: 500824 (C4 (Web Text))

Max Activating Token Index: 547

Text #4

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.2323. Min Act: -0.0000

Data Index: 355129 (C4 (Web Text))

Max Activating Token Index: 334

Text #5

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.1581. Min Act: -0.0000

Data Index: 689174 (C4 (Web Text))

Max Activating Token Index: 94

Text #6

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.1562. Min Act: -0.0000

Data Index: 303598 (C4 (Web Text))

Max Activating Token Index: 880

Text #7

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.0851. Min Act: -0.0000

Data Index: 810076 (C4 (Web Text))

Max Activating Token Index: 207

Text #8

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.8990. Min Act: -0.0000

Data Index: 1522438 (Python Code)

Max Activating Token Index: 473

Text #9

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.8773. Min Act: -0.0000

Data Index: 110322 (C4 (Web Text))

Max Activating Token Index: 838

Text #10

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.0930. Min Act: -0.0000

Data Index: 630126 (C4 (Web Text))

Max Activating Token Index: 697

Text #11

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.8569. Min Act: -0.0000

Data Index: 715468 (C4 (Web Text))

Max Activating Token Index: 126

Text #12

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.8227. Min Act: -0.0000

Data Index: 448309 (C4 (Web Text))

Max Activating Token Index: 655

Text #13

Max Range: 2.5722. Min Range: -2.5722

Max Act: 2.0033. Min Act: -0.0000

Data Index: 1124117 (C4 (Web Text))

Max Activating Token Index: 256

Text #14

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.7858. Min Act: -0.0000

Data Index: 1618324 (Python Code)

Max Activating Token Index: 664

Text #15

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.8600. Min Act: -0.0000

Data Index: 372472 (C4 (Web Text))

Max Activating Token Index: 365

Text #16

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.7492. Min Act: -0.0000

Data Index: 1099186 (C4 (Web Text))

Max Activating Token Index: 311

Text #17

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.8320. Min Act: -0.0000

Data Index: 1083897 (C4 (Web Text))

Max Activating Token Index: 778

Text #18

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.9354. Min Act: -0.0000

Data Index: 834141 (C4 (Web Text))

Max Activating Token Index: 993

Text #19

Max Range: 2.5722. Min Range: -2.5722

Max Act: 1.7902. Min Act: -0.0000

Data Index: 1226528 (C4 (Web Text))

Max Activating Token Index: 892