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

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

Neuron 5162 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4313. Min Range: -3.4313

Max Act: 3.0231. Min Act: -0.0000

Data Index: 1601217 (Python Code)

Max Activating Token Index: 1005

Text #1

Max Range: 3.4313. Min Range: -3.4313

Max Act: 3.3153. Min Act: -0.0000

Data Index: 27087 (C4 (Web Text))

Max Activating Token Index: 765

Text #2

Max Range: 3.4313. Min Range: -3.4313

Max Act: 3.3119. Min Act: -0.0000

Data Index: 1338016 (C4 (Web Text))

Max Activating Token Index: 760

Text #3

Max Range: 3.4313. Min Range: -3.4313

Max Act: 3.4313. Min Act: -0.0000

Data Index: 1589084 (Python Code)

Max Activating Token Index: 953

Text #4

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.9998. Min Act: -0.0000

Data Index: 1306381 (C4 (Web Text))

Max Activating Token Index: 191

Text #5

Max Range: 3.4313. Min Range: -3.4313

Max Act: 3.1548. Min Act: -0.0000

Data Index: 777375 (C4 (Web Text))

Max Activating Token Index: 671

Text #6

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.9366. Min Act: -0.0000

Data Index: 374206 (C4 (Web Text))

Max Activating Token Index: 162

Text #7

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.8935. Min Act: -0.0000

Data Index: 935554 (C4 (Web Text))

Max Activating Token Index: 945

Text #8

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.8723. Min Act: -0.0000

Data Index: 954824 (C4 (Web Text))

Max Activating Token Index: 33

Text #9

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.8550. Min Act: -0.0000

Data Index: 79980 (C4 (Web Text))

Max Activating Token Index: 233

Text #10

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.7822. Min Act: -0.0000

Data Index: 348617 (C4 (Web Text))

Max Activating Token Index: 426

Text #11

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.7851. Min Act: -0.0000

Data Index: 1463448 (Python Code)

Max Activating Token Index: 477

Text #12

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.6230. Min Act: -0.0000

Data Index: 1597614 (Python Code)

Max Activating Token Index: 57

Text #13

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.7307. Min Act: -0.0000

Data Index: 858273 (C4 (Web Text))

Max Activating Token Index: 677

Text #14

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.7006. Min Act: -0.0000

Data Index: 72496 (C4 (Web Text))

Max Activating Token Index: 717

Text #15

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.7010. Min Act: -0.0000

Data Index: 1171507 (C4 (Web Text))

Max Activating Token Index: 59

Text #16

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.5509. Min Act: -0.0000

Data Index: 1184578 (C4 (Web Text))

Max Activating Token Index: 578

Text #17

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.5126. Min Act: -0.0000

Data Index: 1620581 (Python Code)

Max Activating Token Index: 420

Text #18

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.4895. Min Act: -0.0000

Data Index: 341824 (C4 (Web Text))

Max Activating Token Index: 544

Text #19

Max Range: 3.4313. Min Range: -3.4313

Max Act: 2.5968. Min Act: -0.0000

Data Index: 82193 (C4 (Web Text))

Max Activating Token Index: 610