Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1576 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.5734. Min Act: -0.0001

Data Index: 1323564 (C4 (Web Text))

Max Activating Token Index: 638

Text #1

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.5557. Min Act: -0.0001

Data Index: 1599008 (Python Code)

Max Activating Token Index: 420

Text #2

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.5656. Min Act: -0.0001

Data Index: 288018 (C4 (Web Text))

Max Activating Token Index: 783

Text #3

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.5023. Min Act: -0.0001

Data Index: 314535 (C4 (Web Text))

Max Activating Token Index: 674

Text #4

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4899. Min Act: -0.0001

Data Index: 950174 (C4 (Web Text))

Max Activating Token Index: 120

Text #5

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.5334. Min Act: -0.0001

Data Index: 629655 (C4 (Web Text))

Max Activating Token Index: 875

Text #6

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.5170. Min Act: -0.0001

Data Index: 767695 (C4 (Web Text))

Max Activating Token Index: 615

Text #7

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4728. Min Act: -0.0001

Data Index: 62555 (C4 (Web Text))

Max Activating Token Index: 778

Text #8

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.5047. Min Act: -0.0001

Data Index: 646511 (C4 (Web Text))

Max Activating Token Index: 799

Text #9

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.5190. Min Act: -0.0001

Data Index: 24158 (C4 (Web Text))

Max Activating Token Index: 821

Text #10

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4870. Min Act: -0.0001

Data Index: 330095 (C4 (Web Text))

Max Activating Token Index: 842

Text #11

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.5119. Min Act: -0.0001

Data Index: 192372 (C4 (Web Text))

Max Activating Token Index: 681

Text #12

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4927. Min Act: -0.0001

Data Index: 113327 (C4 (Web Text))

Max Activating Token Index: 957

Text #13

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4734. Min Act: -0.0001

Data Index: 1276856 (C4 (Web Text))

Max Activating Token Index: 522

Text #14

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4846. Min Act: -0.0001

Data Index: 1645813 (Python Code)

Max Activating Token Index: 392

Text #15

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4824. Min Act: -0.0001

Data Index: 1635500 (Python Code)

Max Activating Token Index: 61

Text #16

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4794. Min Act: -0.0001

Data Index: 458333 (C4 (Web Text))

Max Activating Token Index: 856

Text #17

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4958. Min Act: -0.0001

Data Index: 80398 (C4 (Web Text))

Max Activating Token Index: 667

Text #18

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4482. Min Act: -0.0001

Data Index: 178121 (C4 (Web Text))

Max Activating Token Index: 560

Text #19

Max Range: 0.5734. Min Range: -0.5734

Max Act: 0.4628. Min Act: -0.0001

Data Index: 235649 (C4 (Web Text))

Max Activating Token Index: 812