Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 3055 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.1314. Min Act: -0.0001

Data Index: 410825 (C4 (Web Text))

Max Activating Token Index: 24

Text #1

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.1091. Min Act: -0.0001

Data Index: 1440310 (Python Code)

Max Activating Token Index: 792

Text #2

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0984. Min Act: 0.0000

Data Index: 1080997 (C4 (Web Text))

Max Activating Token Index: 547

Text #3

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0896. Min Act: 0.0000

Data Index: 1232705 (C4 (Web Text))

Max Activating Token Index: 673

Text #4

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0810. Min Act: -0.0001

Data Index: 1364383 (Python Code)

Max Activating Token Index: 438

Text #5

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0854. Min Act: -0.0001

Data Index: 649509 (C4 (Web Text))

Max Activating Token Index: 811

Text #6

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0851. Min Act: -0.0001

Data Index: 592747 (C4 (Web Text))

Max Activating Token Index: 717

Text #7

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0862. Min Act: 0.0000

Data Index: 959036 (C4 (Web Text))

Max Activating Token Index: 798

Text #8

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0740. Min Act: -0.0001

Data Index: 374473 (C4 (Web Text))

Max Activating Token Index: 554

Text #9

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0753. Min Act: -0.0001

Data Index: 44010 (C4 (Web Text))

Max Activating Token Index: 890

Text #10

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0657. Min Act: -0.0001

Data Index: 1385878 (Python Code)

Max Activating Token Index: 791

Text #11

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0800. Min Act: -0.0001

Data Index: 1151837 (C4 (Web Text))

Max Activating Token Index: 915

Text #12

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0834. Min Act: -0.0001

Data Index: 1242470 (C4 (Web Text))

Max Activating Token Index: 336

Text #13

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0761. Min Act: -0.0000

Data Index: 658239 (C4 (Web Text))

Max Activating Token Index: 90

Text #14

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0806. Min Act: 0.0000

Data Index: 803975 (C4 (Web Text))

Max Activating Token Index: 768

Text #15

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0741. Min Act: -0.0001

Data Index: 1548516 (Python Code)

Max Activating Token Index: 734

Text #16

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0726. Min Act: -0.0001

Data Index: 406270 (C4 (Web Text))

Max Activating Token Index: 576

Text #17

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0738. Min Act: -0.0001

Data Index: 980825 (C4 (Web Text))

Max Activating Token Index: 841

Text #18

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0767. Min Act: -0.0000

Data Index: 48679 (C4 (Web Text))

Max Activating Token Index: 572

Text #19

Max Range: 0.1314. Min Range: -0.1314

Max Act: 0.0611. Min Act: -0.0001

Data Index: 777126 (C4 (Web Text))

Max Activating Token Index: 637