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

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

Neuron 4736 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.7579. Min Act: -0.0000

Data Index: 82313 (C4 (Web Text))

Max Activating Token Index: 529

Text #1

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.7797. Min Act: -0.0000

Data Index: 103554 (C4 (Web Text))

Max Activating Token Index: 319

Text #2

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.7712. Min Act: -0.0000

Data Index: 435605 (C4 (Web Text))

Max Activating Token Index: 361

Text #3

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.6565. Min Act: -0.0000

Data Index: 241402 (C4 (Web Text))

Max Activating Token Index: 980

Text #4

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.6203. Min Act: -0.0000

Data Index: 526377 (C4 (Web Text))

Max Activating Token Index: 49

Text #5

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.6237. Min Act: -0.0000

Data Index: 473211 (C4 (Web Text))

Max Activating Token Index: 346

Text #6

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.7681. Min Act: -0.0000

Data Index: 908160 (C4 (Web Text))

Max Activating Token Index: 695

Text #7

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.6828. Min Act: -0.0000

Data Index: 27435 (C4 (Web Text))

Max Activating Token Index: 191

Text #8

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.6722. Min Act: -0.0000

Data Index: 1572044 (Python Code)

Max Activating Token Index: 254

Text #9

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.6531. Min Act: -0.0000

Data Index: 561372 (C4 (Web Text))

Max Activating Token Index: 161

Text #10

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5667. Min Act: -0.0000

Data Index: 86257 (C4 (Web Text))

Max Activating Token Index: 224

Text #11

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5791. Min Act: -0.0000

Data Index: 1391774 (Python Code)

Max Activating Token Index: 897

Text #12

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5967. Min Act: -0.0000

Data Index: 1401669 (Python Code)

Max Activating Token Index: 788

Text #13

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5092. Min Act: -0.0000

Data Index: 223628 (C4 (Web Text))

Max Activating Token Index: 210

Text #14

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5327. Min Act: -0.0000

Data Index: 1622562 (Python Code)

Max Activating Token Index: 1000

Text #15

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5272. Min Act: -0.0000

Data Index: 1471506 (Python Code)

Max Activating Token Index: 175

Text #16

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5770. Min Act: -0.0000

Data Index: 1639132 (Python Code)

Max Activating Token Index: 428

Text #17

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5562. Min Act: -0.0000

Data Index: 1647771 (Python Code)

Max Activating Token Index: 885

Text #18

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5293. Min Act: -0.0000

Data Index: 1386563 (Python Code)

Max Activating Token Index: 805

Text #19

Max Range: 0.7797. Min Range: -0.7797

Max Act: 0.5593. Min Act: -0.0000

Data Index: 1406630 (Python Code)

Max Activating Token Index: 798