Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2448 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.3712. Min Act: -0.0000

Data Index: 727687 (C4 (Web Text))

Max Activating Token Index: 830

Text #1

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2468. Min Act: -0.0000

Data Index: 1164438 (C4 (Web Text))

Max Activating Token Index: 620

Text #2

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2380. Min Act: -0.0000

Data Index: 158506 (C4 (Web Text))

Max Activating Token Index: 759

Text #3

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2334. Min Act: -0.0000

Data Index: 1089720 (C4 (Web Text))

Max Activating Token Index: 524

Text #4

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2332. Min Act: -0.0000

Data Index: 352812 (C4 (Web Text))

Max Activating Token Index: 558

Text #5

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2336. Min Act: -0.0000

Data Index: 1131755 (C4 (Web Text))

Max Activating Token Index: 804

Text #6

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.1809. Min Act: -0.0000

Data Index: 299420 (C4 (Web Text))

Max Activating Token Index: 934

Text #7

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2481. Min Act: -0.0000

Data Index: 400447 (C4 (Web Text))

Max Activating Token Index: 563

Text #8

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2214. Min Act: -0.0000

Data Index: 544679 (C4 (Web Text))

Max Activating Token Index: 10

Text #9

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2421. Min Act: -0.0000

Data Index: 930353 (C4 (Web Text))

Max Activating Token Index: 390

Text #10

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2166. Min Act: -0.0001

Data Index: 568958 (C4 (Web Text))

Max Activating Token Index: 633

Text #11

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.1838. Min Act: -0.0000

Data Index: 140008 (C4 (Web Text))

Max Activating Token Index: 994

Text #12

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2164. Min Act: -0.0001

Data Index: 1611485 (Python Code)

Max Activating Token Index: 921

Text #13

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2153. Min Act: -0.0001

Data Index: 37242 (C4 (Web Text))

Max Activating Token Index: 506

Text #14

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2114. Min Act: -0.0001

Data Index: 1563729 (Python Code)

Max Activating Token Index: 250

Text #15

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2127. Min Act: -0.0000

Data Index: 278573 (C4 (Web Text))

Max Activating Token Index: 147

Text #16

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2084. Min Act: -0.0001

Data Index: 1549609 (Python Code)

Max Activating Token Index: 1011

Text #17

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2115. Min Act: -0.0000

Data Index: 1176694 (C4 (Web Text))

Max Activating Token Index: 203

Text #18

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2068. Min Act: -0.0000

Data Index: 1049203 (C4 (Web Text))

Max Activating Token Index: 273

Text #19

Max Range: 0.3712. Min Range: -0.3712

Max Act: 0.2057. Min Act: -0.0001

Data Index: 332921 (C4 (Web Text))

Max Activating Token Index: 189