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

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

Neuron 1057 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.7595. Min Act: -0.0000

Data Index: 1508675 (Python Code)

Max Activating Token Index: 916

Text #1

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.7356. Min Act: -0.0000

Data Index: 952667 (C4 (Web Text))

Max Activating Token Index: 288

Text #2

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.6668. Min Act: -0.0000

Data Index: 842836 (C4 (Web Text))

Max Activating Token Index: 723

Text #3

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.6593. Min Act: -0.0000

Data Index: 158920 (C4 (Web Text))

Max Activating Token Index: 38

Text #4

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.5650. Min Act: -0.0000

Data Index: 911953 (C4 (Web Text))

Max Activating Token Index: 290

Text #5

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.7137. Min Act: -0.0000

Data Index: 1517092 (Python Code)

Max Activating Token Index: 257

Text #6

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.6331. Min Act: -0.0000

Data Index: 1101869 (C4 (Web Text))

Max Activating Token Index: 736

Text #7

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.6757. Min Act: -0.0000

Data Index: 1442411 (Python Code)

Max Activating Token Index: 822

Text #8

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.6465. Min Act: -0.0000

Data Index: 1234389 (C4 (Web Text))

Max Activating Token Index: 123

Text #9

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.5242. Min Act: -0.0000

Data Index: 1340565 (C4 (Web Text))

Max Activating Token Index: 841

Text #10

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.6955. Min Act: -0.0000

Data Index: 374972 (C4 (Web Text))

Max Activating Token Index: 765

Text #11

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.5796. Min Act: -0.0000

Data Index: 1367203 (Python Code)

Max Activating Token Index: 990

Text #12

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.6710. Min Act: -0.0000

Data Index: 728168 (C4 (Web Text))

Max Activating Token Index: 882

Text #13

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.5985. Min Act: -0.0000

Data Index: 742529 (C4 (Web Text))

Max Activating Token Index: 188

Text #14

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.5891. Min Act: -0.0000

Data Index: 577200 (C4 (Web Text))

Max Activating Token Index: 456

Text #15

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.5668. Min Act: -0.0000

Data Index: 648162 (C4 (Web Text))

Max Activating Token Index: 10

Text #16

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.6412. Min Act: -0.0000

Data Index: 384278 (C4 (Web Text))

Max Activating Token Index: 228

Text #17

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.6553. Min Act: -0.0000

Data Index: 1388667 (Python Code)

Max Activating Token Index: 293

Text #18

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.5714. Min Act: -0.0000

Data Index: 940858 (C4 (Web Text))

Max Activating Token Index: 327

Text #19

Max Range: 0.7595. Min Range: -0.7595

Max Act: 0.5681. Min Act: -0.0000

Data Index: 965606 (C4 (Web Text))

Max Activating Token Index: 103