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

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

Neuron 1600 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.9650. Min Act: -0.0000

Data Index: 1033889 (C4 (Web Text))

Max Activating Token Index: 593

Text #1

Max Range: 1.0050. Min Range: -1.0050

Max Act: 1.0050. Min Act: -0.0000

Data Index: 800525 (C4 (Web Text))

Max Activating Token Index: 316

Text #2

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.9776. Min Act: -0.0000

Data Index: 1070147 (C4 (Web Text))

Max Activating Token Index: 325

Text #3

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.9867. Min Act: -0.0000

Data Index: 764696 (C4 (Web Text))

Max Activating Token Index: 92

Text #4

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.9572. Min Act: -0.0000

Data Index: 862146 (C4 (Web Text))

Max Activating Token Index: 489

Text #5

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.9728. Min Act: -0.0000

Data Index: 1167632 (C4 (Web Text))

Max Activating Token Index: 611

Text #6

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.8899. Min Act: -0.0000

Data Index: 1656728 (Python Code)

Max Activating Token Index: 590

Text #7

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.8400. Min Act: -0.0000

Data Index: 1367961 (Python Code)

Max Activating Token Index: 794

Text #8

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.7746. Min Act: -0.0000

Data Index: 1375835 (Python Code)

Max Activating Token Index: 734

Text #9

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.8363. Min Act: -0.0000

Data Index: 1495515 (Python Code)

Max Activating Token Index: 807

Text #10

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.7470. Min Act: -0.0000

Data Index: 1335747 (C4 (Web Text))

Max Activating Token Index: 686

Text #11

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.7234. Min Act: -0.0000

Data Index: 905982 (C4 (Web Text))

Max Activating Token Index: 371

Text #12

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.8621. Min Act: -0.0000

Data Index: 1416777 (Python Code)

Max Activating Token Index: 868

Text #13

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.8177. Min Act: 0.0000

Data Index: 1474277 (Python Code)

Max Activating Token Index: 753

Text #14

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.8211. Min Act: -0.0000

Data Index: 1406479 (Python Code)

Max Activating Token Index: 486

Text #15

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.7594. Min Act: -0.0000

Data Index: 383745 (C4 (Web Text))

Max Activating Token Index: 235

Text #16

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.8563. Min Act: -0.0000

Data Index: 1200692 (C4 (Web Text))

Max Activating Token Index: 641

Text #17

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.7266. Min Act: -0.0000

Data Index: 185393 (C4 (Web Text))

Max Activating Token Index: 716

Text #18

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.7531. Min Act: -0.0000

Data Index: 606145 (C4 (Web Text))

Max Activating Token Index: 841

Text #19

Max Range: 1.0050. Min Range: -1.0050

Max Act: 0.7648. Min Act: -0.0000

Data Index: 1589870 (Python Code)

Max Activating Token Index: 716