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

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

Neuron 4837 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7904. Min Range: -2.7904

Max Act: 2.7904. Min Act: 0.0000

Data Index: 53532 (C4 (Web Text))

Max Activating Token Index: 128

Text #1

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.7827. Min Act: -0.0000

Data Index: 625368 (C4 (Web Text))

Max Activating Token Index: 643

Text #2

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.9141. Min Act: -0.0000

Data Index: 1475350 (Python Code)

Max Activating Token Index: 701

Text #3

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.7209. Min Act: -0.0000

Data Index: 1634471 (Python Code)

Max Activating Token Index: 888

Text #4

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.7142. Min Act: -0.0000

Data Index: 1556099 (Python Code)

Max Activating Token Index: 259

Text #5

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.8289. Min Act: -0.0000

Data Index: 1087396 (C4 (Web Text))

Max Activating Token Index: 982

Text #6

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.6278. Min Act: -0.0000

Data Index: 527143 (C4 (Web Text))

Max Activating Token Index: 985

Text #7

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.6837. Min Act: -0.0000

Data Index: 530079 (C4 (Web Text))

Max Activating Token Index: 735

Text #8

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.6233. Min Act: -0.0000

Data Index: 1297940 (C4 (Web Text))

Max Activating Token Index: 597

Text #9

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.5084. Min Act: -0.0000

Data Index: 56607 (C4 (Web Text))

Max Activating Token Index: 457

Text #10

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.5767. Min Act: -0.0000

Data Index: 1087733 (C4 (Web Text))

Max Activating Token Index: 138

Text #11

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.4466. Min Act: -0.0000

Data Index: 1372114 (Python Code)

Max Activating Token Index: 488

Text #12

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.5693. Min Act: -0.0000

Data Index: 1387306 (Python Code)

Max Activating Token Index: 842

Text #13

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.5168. Min Act: -0.0000

Data Index: 1418822 (Python Code)

Max Activating Token Index: 378

Text #14

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.4653. Min Act: -0.0000

Data Index: 1642835 (Python Code)

Max Activating Token Index: 547

Text #15

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.4963. Min Act: -0.0000

Data Index: 1593423 (Python Code)

Max Activating Token Index: 879

Text #16

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.4545. Min Act: -0.0000

Data Index: 1402843 (Python Code)

Max Activating Token Index: 928

Text #17

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.3441. Min Act: -0.0000

Data Index: 1594527 (Python Code)

Max Activating Token Index: 127

Text #18

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.4984. Min Act: -0.0000

Data Index: 763241 (C4 (Web Text))

Max Activating Token Index: 385

Text #19

Max Range: 2.7904. Min Range: -2.7904

Max Act: 1.3093. Min Act: -0.0000

Data Index: 1546859 (Python Code)

Max Activating Token Index: 951