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

Dataset: The Pile

Neuron 1053 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.1773. Min Act: -0.0000

Data Index: 1435176 (The Pile)

Max Activating Token Index: 963

Text #1

Max Range: 8.1560. Min Range: -8.1560

Max Act: 8.1560. Min Act: -0.0000

Data Index: 639199 (The Pile)

Max Activating Token Index: 128

Text #2

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.0425. Min Act: -0.0000

Data Index: 829723 (The Pile)

Max Activating Token Index: 665

Text #3

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.2091. Min Act: -0.0000

Data Index: 1838448 (The Pile)

Max Activating Token Index: 551

Text #4

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.3992. Min Act: -0.0000

Data Index: 1265094 (The Pile)

Max Activating Token Index: 532

Text #5

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.1341. Min Act: -0.0000

Data Index: 374045 (The Pile)

Max Activating Token Index: 271

Text #6

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.2947. Min Act: -0.0000

Data Index: 1161053 (The Pile)

Max Activating Token Index: 837

Text #7

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.0409. Min Act: -0.0000

Data Index: 1690614 (The Pile)

Max Activating Token Index: 544

Text #8

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.0179. Min Act: -0.0000

Data Index: 1912381 (The Pile)

Max Activating Token Index: 350

Text #9

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.4254. Min Act: -0.0000

Data Index: 407583 (The Pile)

Max Activating Token Index: 108

Text #10

Max Range: 8.1560. Min Range: -8.1560

Max Act: 6.9998. Min Act: -0.0000

Data Index: 618986 (The Pile)

Max Activating Token Index: 386

Text #11

Max Range: 8.1560. Min Range: -8.1560

Max Act: 6.9272. Min Act: -0.0000

Data Index: 1535173 (The Pile)

Max Activating Token Index: 927

Text #12

Max Range: 8.1560. Min Range: -8.1560

Max Act: 6.4826. Min Act: -0.0000

Data Index: 1693340 (The Pile)

Max Activating Token Index: 766

Text #13

Max Range: 8.1560. Min Range: -8.1560

Max Act: 6.2904. Min Act: -0.0000

Data Index: 353170 (The Pile)

Max Activating Token Index: 530

Text #14

Max Range: 8.1560. Min Range: -8.1560

Max Act: 6.8028. Min Act: -0.0000

Data Index: 581276 (The Pile)

Max Activating Token Index: 150

Text #15

Max Range: 8.1560. Min Range: -8.1560

Max Act: 6.6065. Min Act: -0.0000

Data Index: 1865172 (The Pile)

Max Activating Token Index: 199

Text #16

Max Range: 8.1560. Min Range: -8.1560

Max Act: 6.7577. Min Act: -0.0000

Data Index: 515632 (The Pile)

Max Activating Token Index: 472

Text #17

Max Range: 8.1560. Min Range: -8.1560

Max Act: 6.5432. Min Act: -0.0000

Data Index: 670936 (The Pile)

Max Activating Token Index: 452

Text #18

Max Range: 8.1560. Min Range: -8.1560

Max Act: 5.7092. Min Act: -0.0000

Data Index: 1524320 (The Pile)

Max Activating Token Index: 1009

Text #19

Max Range: 8.1560. Min Range: -8.1560

Max Act: 7.1587. Min Act: -0.0000

Data Index: 1015343 (The Pile)

Max Activating Token Index: 87