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

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

Neuron 3977 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.7648. Min Act: -0.0001

Data Index: 239522 (C4 (Web Text))

Max Activating Token Index: 436

Text #1

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.8004. Min Act: -0.0001

Data Index: 941312 (C4 (Web Text))

Max Activating Token Index: 545

Text #2

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.7544. Min Act: -0.0001

Data Index: 286730 (C4 (Web Text))

Max Activating Token Index: 974

Text #3

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.7879. Min Act: -0.0001

Data Index: 1089140 (C4 (Web Text))

Max Activating Token Index: 678

Text #4

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.7354. Min Act: -0.0001

Data Index: 133751 (C4 (Web Text))

Max Activating Token Index: 747

Text #5

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.7276. Min Act: -0.0001

Data Index: 1325419 (C4 (Web Text))

Max Activating Token Index: 420

Text #6

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.7252. Min Act: -0.0001

Data Index: 674958 (C4 (Web Text))

Max Activating Token Index: 948

Text #7

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6712. Min Act: -0.0001

Data Index: 1628351 (Python Code)

Max Activating Token Index: 438

Text #8

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6743. Min Act: -0.0001

Data Index: 898847 (C4 (Web Text))

Max Activating Token Index: 1014

Text #9

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6968. Min Act: -0.0001

Data Index: 128672 (C4 (Web Text))

Max Activating Token Index: 964

Text #10

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6604. Min Act: -0.0001

Data Index: 209618 (C4 (Web Text))

Max Activating Token Index: 449

Text #11

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6675. Min Act: -0.0001

Data Index: 1382521 (Python Code)

Max Activating Token Index: 877

Text #12

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6689. Min Act: -0.0001

Data Index: 980644 (C4 (Web Text))

Max Activating Token Index: 963

Text #13

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6665. Min Act: -0.0001

Data Index: 934204 (C4 (Web Text))

Max Activating Token Index: 999

Text #14

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6660. Min Act: -0.0001

Data Index: 134217 (C4 (Web Text))

Max Activating Token Index: 384

Text #15

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6622. Min Act: -0.0001

Data Index: 1597462 (Python Code)

Max Activating Token Index: 702

Text #16

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.5885. Min Act: -0.0001

Data Index: 1481797 (Python Code)

Max Activating Token Index: 798

Text #17

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6724. Min Act: -0.0001

Data Index: 261890 (C4 (Web Text))

Max Activating Token Index: 497

Text #18

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6368. Min Act: -0.0001

Data Index: 423043 (C4 (Web Text))

Max Activating Token Index: 333

Text #19

Max Range: 0.8004. Min Range: -0.8004

Max Act: 0.6284. Min Act: -0.0001

Data Index: 184633 (C4 (Web Text))

Max Activating Token Index: 459