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

Dataset: The Pile

Neuron 1690 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: 8.2065. Min Range: -8.2065

Max Act: 8.2065. Min Act: -0.0000

Data Index: 1349291 (The Pile)

Max Activating Token Index: 62

Text #1

Max Range: 8.2065. Min Range: -8.2065

Max Act: 8.1536. Min Act: -0.0000

Data Index: 1953120 (The Pile)

Max Activating Token Index: 185

Text #2

Max Range: 8.2065. Min Range: -8.2065

Max Act: 7.2969. Min Act: -0.0000

Data Index: 1125011 (The Pile)

Max Activating Token Index: 460

Text #3

Max Range: 8.2065. Min Range: -8.2065

Max Act: 7.7627. Min Act: -0.0000

Data Index: 488141 (The Pile)

Max Activating Token Index: 185

Text #4

Max Range: 8.2065. Min Range: -8.2065

Max Act: 7.2420. Min Act: -0.0000

Data Index: 951234 (The Pile)

Max Activating Token Index: 48

Text #5

Max Range: 8.2065. Min Range: -8.2065

Max Act: 7.8636. Min Act: -0.0000

Data Index: 1959083 (The Pile)

Max Activating Token Index: 714

Text #6

Max Range: 8.2065. Min Range: -8.2065

Max Act: 7.5944. Min Act: -0.0000

Data Index: 1007365 (The Pile)

Max Activating Token Index: 239

Text #7

Max Range: 8.2065. Min Range: -8.2065

Max Act: 6.5985. Min Act: -0.0000

Data Index: 276001 (The Pile)

Max Activating Token Index: 435

Text #8

Max Range: 8.2065. Min Range: -8.2065

Max Act: 6.9622. Min Act: -0.0000

Data Index: 1837614 (The Pile)

Max Activating Token Index: 98

Text #9

Max Range: 8.2065. Min Range: -8.2065

Max Act: 6.3215. Min Act: -0.0000

Data Index: 277050 (The Pile)

Max Activating Token Index: 123

Text #10

Max Range: 8.2065. Min Range: -8.2065

Max Act: 7.1738. Min Act: -0.0000

Data Index: 185345 (The Pile)

Max Activating Token Index: 427

Text #11

Max Range: 8.2065. Min Range: -8.2065

Max Act: 6.8020. Min Act: -0.0000

Data Index: 342733 (The Pile)

Max Activating Token Index: 486

Text #12

Max Range: 8.2065. Min Range: -8.2065

Max Act: 6.8011. Min Act: -0.0000

Data Index: 1865999 (The Pile)

Max Activating Token Index: 76

Text #13

Max Range: 8.2065. Min Range: -8.2065

Max Act: 6.3520. Min Act: -0.0000

Data Index: 53968 (The Pile)

Max Activating Token Index: 210

Text #14

Max Range: 8.2065. Min Range: -8.2065

Max Act: 6.6434. Min Act: -0.0000

Data Index: 598665 (The Pile)

Max Activating Token Index: 656

Text #15

Max Range: 8.2065. Min Range: -8.2065

Max Act: 6.3183. Min Act: -0.0000

Data Index: 1379566 (The Pile)

Max Activating Token Index: 529

Text #16

Max Range: 8.2065. Min Range: -8.2065

Max Act: 5.5812. Min Act: -0.0000

Data Index: 441547 (The Pile)

Max Activating Token Index: 410

Text #17

Max Range: 8.2065. Min Range: -8.2065

Max Act: 5.5112. Min Act: -0.0000

Data Index: 568593 (The Pile)

Max Activating Token Index: 181

Text #18

Max Range: 8.2065. Min Range: -8.2065

Max Act: 5.7947. Min Act: -0.0000

Data Index: 1709293 (The Pile)

Max Activating Token Index: 335

Text #19

Max Range: 8.2065. Min Range: -8.2065

Max Act: 6.3127. Min Act: -0.0000

Data Index: 1484424 (The Pile)

Max Activating Token Index: 539