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

Dataset: The Pile

Neuron 2244 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.5052. Min Act: -0.0000

Data Index: 518592 (The Pile)

Max Activating Token Index: 632

Text #1

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.2829. Min Act: -0.0000

Data Index: 77555 (The Pile)

Max Activating Token Index: 522

Text #2

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.5241. Min Act: -0.0000

Data Index: 1019143 (The Pile)

Max Activating Token Index: 554

Text #3

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.0792. Min Act: -0.0000

Data Index: 1151897 (The Pile)

Max Activating Token Index: 409

Text #4

Max Range: 8.5241. Min Range: -8.5241

Max Act: 7.9160. Min Act: -0.0000

Data Index: 1182447 (The Pile)

Max Activating Token Index: 598

Text #5

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.5033. Min Act: -0.0000

Data Index: 985738 (The Pile)

Max Activating Token Index: 568

Text #6

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.1650. Min Act: -0.0000

Data Index: 1751702 (The Pile)

Max Activating Token Index: 480

Text #7

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.1535. Min Act: -0.0000

Data Index: 1981315 (The Pile)

Max Activating Token Index: 594

Text #8

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.0547. Min Act: -0.0000

Data Index: 729862 (The Pile)

Max Activating Token Index: 610

Text #9

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.2529. Min Act: -0.0000

Data Index: 714258 (The Pile)

Max Activating Token Index: 553

Text #10

Max Range: 8.5241. Min Range: -8.5241

Max Act: 7.6599. Min Act: -0.0000

Data Index: 583331 (The Pile)

Max Activating Token Index: 430

Text #11

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.2304. Min Act: -0.0000

Data Index: 1318999 (The Pile)

Max Activating Token Index: 581

Text #12

Max Range: 8.5241. Min Range: -8.5241

Max Act: 7.9709. Min Act: -0.0000

Data Index: 924023 (The Pile)

Max Activating Token Index: 511

Text #13

Max Range: 8.5241. Min Range: -8.5241

Max Act: 8.3892. Min Act: -0.0000

Data Index: 1288390 (The Pile)

Max Activating Token Index: 586

Text #14

Max Range: 8.5241. Min Range: -8.5241

Max Act: 7.4911. Min Act: -0.0000

Data Index: 816104 (The Pile)

Max Activating Token Index: 531

Text #15

Max Range: 8.5241. Min Range: -8.5241

Max Act: 7.9663. Min Act: -0.0000

Data Index: 1664252 (The Pile)

Max Activating Token Index: 623

Text #16

Max Range: 8.5241. Min Range: -8.5241

Max Act: 7.8356. Min Act: -0.0000

Data Index: 967534 (The Pile)

Max Activating Token Index: 556

Text #17

Max Range: 8.5241. Min Range: -8.5241

Max Act: 7.2056. Min Act: -0.0000

Data Index: 395966 (The Pile)

Max Activating Token Index: 343

Text #18

Max Range: 8.5241. Min Range: -8.5241

Max Act: 7.8386. Min Act: -0.0000

Data Index: 1637358 (The Pile)

Max Activating Token Index: 446

Text #19

Max Range: 8.5241. Min Range: -8.5241

Max Act: 7.7108. Min Act: -0.0000

Data Index: 515826 (The Pile)

Max Activating Token Index: 399