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

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

Neuron 5769 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.5732. Min Act: -0.0002

Data Index: 1378150 (Python Code)

Max Activating Token Index: 643

Text #1

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4968. Min Act: -0.0002

Data Index: 517826 (C4 (Web Text))

Max Activating Token Index: 879

Text #2

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4727. Min Act: -0.0002

Data Index: 1416813 (Python Code)

Max Activating Token Index: 722

Text #3

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4428. Min Act: -0.0002

Data Index: 32051 (C4 (Web Text))

Max Activating Token Index: 105

Text #4

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.3832. Min Act: -0.0002

Data Index: 234533 (C4 (Web Text))

Max Activating Token Index: 363

Text #5

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4498. Min Act: -0.0002

Data Index: 382948 (C4 (Web Text))

Max Activating Token Index: 380

Text #6

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4568. Min Act: -0.0002

Data Index: 419946 (C4 (Web Text))

Max Activating Token Index: 970

Text #7

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4507. Min Act: -0.0002

Data Index: 1007484 (C4 (Web Text))

Max Activating Token Index: 301

Text #8

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.3925. Min Act: -0.0002

Data Index: 836896 (C4 (Web Text))

Max Activating Token Index: 767

Text #9

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.3930. Min Act: -0.0002

Data Index: 397743 (C4 (Web Text))

Max Activating Token Index: 255

Text #10

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4042. Min Act: -0.0002

Data Index: 112541 (C4 (Web Text))

Max Activating Token Index: 166

Text #11

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4045. Min Act: -0.0002

Data Index: 1202335 (C4 (Web Text))

Max Activating Token Index: 131

Text #12

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4022. Min Act: -0.0002

Data Index: 68029 (C4 (Web Text))

Max Activating Token Index: 736

Text #13

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.3670. Min Act: -0.0002

Data Index: 973452 (C4 (Web Text))

Max Activating Token Index: 214

Text #14

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.3930. Min Act: -0.0002

Data Index: 708867 (C4 (Web Text))

Max Activating Token Index: 303

Text #15

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.3894. Min Act: -0.0002

Data Index: 1373374 (Python Code)

Max Activating Token Index: 795

Text #16

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4079. Min Act: -0.0002

Data Index: 274975 (C4 (Web Text))

Max Activating Token Index: 106

Text #17

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.4072. Min Act: -0.0002

Data Index: 514183 (C4 (Web Text))

Max Activating Token Index: 338

Text #18

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.3750. Min Act: -0.0002

Data Index: 1470076 (Python Code)

Max Activating Token Index: 744

Text #19

Max Range: 0.5732. Min Range: -0.5732

Max Act: 0.3732. Min Act: -0.0002

Data Index: 1541634 (Python Code)

Max Activating Token Index: 339