Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2091 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.6093. Min Act: -0.0001

Data Index: 223805 (C4 (Web Text))

Max Activating Token Index: 655

Text #1

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.4279. Min Act: -0.0001

Data Index: 925310 (C4 (Web Text))

Max Activating Token Index: 965

Text #2

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.4447. Min Act: -0.0001

Data Index: 951402 (C4 (Web Text))

Max Activating Token Index: 377

Text #3

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.3494. Min Act: -0.0001

Data Index: 475631 (C4 (Web Text))

Max Activating Token Index: 663

Text #4

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.1257. Min Act: -0.0001

Data Index: 985657 (C4 (Web Text))

Max Activating Token Index: 892

Text #5

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.1378. Min Act: -0.0001

Data Index: 13490 (C4 (Web Text))

Max Activating Token Index: 579

Text #6

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.0916. Min Act: -0.0001

Data Index: 620176 (C4 (Web Text))

Max Activating Token Index: 80

Text #7

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.0829. Min Act: -0.0001

Data Index: 1301844 (C4 (Web Text))

Max Activating Token Index: 171

Text #8

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.0710. Min Act: -0.0001

Data Index: 126488 (C4 (Web Text))

Max Activating Token Index: 674

Text #9

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9998. Min Act: -0.0001

Data Index: 1129733 (C4 (Web Text))

Max Activating Token Index: 668

Text #10

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.0405. Min Act: -0.0001

Data Index: 1557430 (Python Code)

Max Activating Token Index: 470

Text #11

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9865. Min Act: -0.0001

Data Index: 160698 (C4 (Web Text))

Max Activating Token Index: 540

Text #12

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.0006. Min Act: -0.0001

Data Index: 1414433 (Python Code)

Max Activating Token Index: 138

Text #13

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9918. Min Act: -0.0001

Data Index: 1068208 (C4 (Web Text))

Max Activating Token Index: 834

Text #14

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.0670. Min Act: -0.0001

Data Index: 221633 (C4 (Web Text))

Max Activating Token Index: 279

Text #15

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9476. Min Act: -0.0001

Data Index: 702908 (C4 (Web Text))

Max Activating Token Index: 194

Text #16

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9116. Min Act: -0.0001

Data Index: 964761 (C4 (Web Text))

Max Activating Token Index: 573

Text #17

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9098. Min Act: -0.0001

Data Index: 40218 (C4 (Web Text))

Max Activating Token Index: 615

Text #18

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9657. Min Act: -0.0001

Data Index: 1622030 (Python Code)

Max Activating Token Index: 788

Text #19

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.8732. Min Act: -0.0001

Data Index: 963467 (C4 (Web Text))

Max Activating Token Index: 336