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

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

Neuron 2112 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.9882. Min Act: -0.0001

Data Index: 748509 (C4 (Web Text))

Max Activating Token Index: 637

Text #1

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.7955. Min Act: -0.0001

Data Index: 156943 (C4 (Web Text))

Max Activating Token Index: 594

Text #2

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.7813. Min Act: -0.0001

Data Index: 642166 (C4 (Web Text))

Max Activating Token Index: 25

Text #3

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.9770. Min Act: -0.0001

Data Index: 80825 (C4 (Web Text))

Max Activating Token Index: 162

Text #4

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.5636. Min Act: -0.0001

Data Index: 172568 (C4 (Web Text))

Max Activating Token Index: 82

Text #5

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.8793. Min Act: -0.0001

Data Index: 1507135 (Python Code)

Max Activating Token Index: 533

Text #6

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.7451. Min Act: -0.0001

Data Index: 878328 (C4 (Web Text))

Max Activating Token Index: 601

Text #7

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.9769. Min Act: -0.0001

Data Index: 1152964 (C4 (Web Text))

Max Activating Token Index: 338

Text #8

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.6607. Min Act: -0.0001

Data Index: 256005 (C4 (Web Text))

Max Activating Token Index: 691

Text #9

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.6971. Min Act: -0.0001

Data Index: 91147 (C4 (Web Text))

Max Activating Token Index: 452

Text #10

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.7266. Min Act: -0.0001

Data Index: 643127 (C4 (Web Text))

Max Activating Token Index: 669

Text #11

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.6822. Min Act: -0.0001

Data Index: 1299402 (C4 (Web Text))

Max Activating Token Index: 303

Text #12

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.4957. Min Act: -0.0001

Data Index: 51023 (C4 (Web Text))

Max Activating Token Index: 931

Text #13

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.6182. Min Act: -0.0001

Data Index: 24710 (C4 (Web Text))

Max Activating Token Index: 350

Text #14

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.4800. Min Act: -0.0001

Data Index: 392664 (C4 (Web Text))

Max Activating Token Index: 156

Text #15

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.5950. Min Act: -0.0001

Data Index: 1214811 (C4 (Web Text))

Max Activating Token Index: 434

Text #16

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.5088. Min Act: -0.0001

Data Index: 1193682 (C4 (Web Text))

Max Activating Token Index: 254

Text #17

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.5782. Min Act: -0.0001

Data Index: 264850 (C4 (Web Text))

Max Activating Token Index: 31

Text #18

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.6207. Min Act: -0.0001

Data Index: 727026 (C4 (Web Text))

Max Activating Token Index: 413

Text #19

Max Range: 1.9882. Min Range: -1.9882

Max Act: 1.6592. Min Act: -0.0001

Data Index: 1084575 (C4 (Web Text))

Max Activating Token Index: 447