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

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

Neuron 3705 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5970. Min Range: -3.5970

Max Act: 3.5970. Min Act: -0.0000

Data Index: 753996 (C4 (Web Text))

Max Activating Token Index: 310

Text #1

Max Range: 3.5970. Min Range: -3.5970

Max Act: 3.3007. Min Act: -0.0000

Data Index: 1277654 (C4 (Web Text))

Max Activating Token Index: 698

Text #2

Max Range: 3.5970. Min Range: -3.5970

Max Act: 3.0092. Min Act: -0.0000

Data Index: 1384884 (Python Code)

Max Activating Token Index: 247

Text #3

Max Range: 3.5970. Min Range: -3.5970

Max Act: 3.0214. Min Act: -0.0000

Data Index: 516245 (C4 (Web Text))

Max Activating Token Index: 896

Text #4

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.9457. Min Act: -0.0000

Data Index: 895168 (C4 (Web Text))

Max Activating Token Index: 636

Text #5

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.8837. Min Act: -0.0001

Data Index: 1150157 (C4 (Web Text))

Max Activating Token Index: 127

Text #6

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.6649. Min Act: -0.0000

Data Index: 1363861 (Python Code)

Max Activating Token Index: 13

Text #7

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.8045. Min Act: -0.0000

Data Index: 1490320 (Python Code)

Max Activating Token Index: 228

Text #8

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.6363. Min Act: -0.0000

Data Index: 1580451 (Python Code)

Max Activating Token Index: 99

Text #9

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.7610. Min Act: -0.0000

Data Index: 1384848 (Python Code)

Max Activating Token Index: 50

Text #10

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.7238. Min Act: -0.0000

Data Index: 1443773 (Python Code)

Max Activating Token Index: 23

Text #11

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.5691. Min Act: -0.0000

Data Index: 1379772 (Python Code)

Max Activating Token Index: 588

Text #12

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.5440. Min Act: -0.0000

Data Index: 1336250 (C4 (Web Text))

Max Activating Token Index: 173

Text #13

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.6196. Min Act: -0.0000

Data Index: 37775 (C4 (Web Text))

Max Activating Token Index: 584

Text #14

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.5370. Min Act: -0.0000

Data Index: 238418 (C4 (Web Text))

Max Activating Token Index: 850

Text #15

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.5387. Min Act: -0.0000

Data Index: 1468360 (Python Code)

Max Activating Token Index: 837

Text #16

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.4097. Min Act: -0.0000

Data Index: 1443450 (Python Code)

Max Activating Token Index: 816

Text #17

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.7714. Min Act: -0.0000

Data Index: 310683 (C4 (Web Text))

Max Activating Token Index: 973

Text #18

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.4958. Min Act: -0.0000

Data Index: 1406191 (Python Code)

Max Activating Token Index: 580

Text #19

Max Range: 3.5970. Min Range: -3.5970

Max Act: 2.4824. Min Act: -0.0000

Data Index: 1508856 (Python Code)

Max Activating Token Index: 570