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

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

Neuron 2047 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.7288. Min Act: -0.0001

Data Index: 722681 (C4 (Web Text))

Max Activating Token Index: 657

Text #1

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.6742. Min Act: -0.0001

Data Index: 640777 (C4 (Web Text))

Max Activating Token Index: 587

Text #2

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.6961. Min Act: -0.0001

Data Index: 1464367 (Python Code)

Max Activating Token Index: 102

Text #3

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.6270. Min Act: -0.0001

Data Index: 824620 (C4 (Web Text))

Max Activating Token Index: 741

Text #4

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5933. Min Act: -0.0001

Data Index: 1110178 (C4 (Web Text))

Max Activating Token Index: 262

Text #5

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.6310. Min Act: -0.0001

Data Index: 538268 (C4 (Web Text))

Max Activating Token Index: 85

Text #6

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5812. Min Act: -0.0001

Data Index: 882951 (C4 (Web Text))

Max Activating Token Index: 480

Text #7

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5410. Min Act: -0.0001

Data Index: 329344 (C4 (Web Text))

Max Activating Token Index: 345

Text #8

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5329. Min Act: -0.0001

Data Index: 1032088 (C4 (Web Text))

Max Activating Token Index: 423

Text #9

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5238. Min Act: -0.0001

Data Index: 478275 (C4 (Web Text))

Max Activating Token Index: 385

Text #10

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5166. Min Act: -0.0001

Data Index: 318691 (C4 (Web Text))

Max Activating Token Index: 892

Text #11

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5084. Min Act: -0.0001

Data Index: 1461132 (Python Code)

Max Activating Token Index: 33

Text #12

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5382. Min Act: -0.0001

Data Index: 1618229 (Python Code)

Max Activating Token Index: 156

Text #13

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5049. Min Act: -0.0001

Data Index: 1107165 (C4 (Web Text))

Max Activating Token Index: 928

Text #14

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5624. Min Act: -0.0001

Data Index: 1450670 (Python Code)

Max Activating Token Index: 533

Text #15

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.4716. Min Act: -0.0001

Data Index: 565117 (C4 (Web Text))

Max Activating Token Index: 410

Text #16

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.4972. Min Act: -0.0001

Data Index: 479298 (C4 (Web Text))

Max Activating Token Index: 4

Text #17

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5115. Min Act: -0.0001

Data Index: 811858 (C4 (Web Text))

Max Activating Token Index: 12

Text #18

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5104. Min Act: -0.0001

Data Index: 1346272 (C4 (Web Text))

Max Activating Token Index: 38

Text #19

Max Range: 0.7288. Min Range: -0.7288

Max Act: 0.5008. Min Act: -0.0001

Data Index: 890749 (C4 (Web Text))

Max Activating Token Index: 756