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

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

Neuron 2640 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5086. Min Range: -3.5086

Max Act: 3.5086. Min Act: -0.0000

Data Index: 692464 (C4 (Web Text))

Max Activating Token Index: 22

Text #1

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.9574. Min Act: -0.0000

Data Index: 492350 (C4 (Web Text))

Max Activating Token Index: 755

Text #2

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.9021. Min Act: -0.0000

Data Index: 806615 (C4 (Web Text))

Max Activating Token Index: 15

Text #3

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.8730. Min Act: -0.0000

Data Index: 1177681 (C4 (Web Text))

Max Activating Token Index: 24

Text #4

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.8682. Min Act: -0.0000

Data Index: 1084610 (C4 (Web Text))

Max Activating Token Index: 161

Text #5

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.6104. Min Act: -0.0000

Data Index: 1354188 (C4 (Web Text))

Max Activating Token Index: 77

Text #6

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.6026. Min Act: -0.0000

Data Index: 562571 (C4 (Web Text))

Max Activating Token Index: 79

Text #7

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.5786. Min Act: -0.0000

Data Index: 132096 (C4 (Web Text))

Max Activating Token Index: 218

Text #8

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.7294. Min Act: -0.0000

Data Index: 433879 (C4 (Web Text))

Max Activating Token Index: 229

Text #9

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.5463. Min Act: -0.0000

Data Index: 1608541 (Python Code)

Max Activating Token Index: 729

Text #10

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.7954. Min Act: -0.0000

Data Index: 717330 (C4 (Web Text))

Max Activating Token Index: 187

Text #11

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.6541. Min Act: -0.0000

Data Index: 484406 (C4 (Web Text))

Max Activating Token Index: 341

Text #12

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.6597. Min Act: -0.0000

Data Index: 770195 (C4 (Web Text))

Max Activating Token Index: 22

Text #13

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.9272. Min Act: -0.0000

Data Index: 1181771 (C4 (Web Text))

Max Activating Token Index: 904

Text #14

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.4875. Min Act: -0.0000

Data Index: 458936 (C4 (Web Text))

Max Activating Token Index: 661

Text #15

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.6111. Min Act: -0.0000

Data Index: 1357048 (C4 (Web Text))

Max Activating Token Index: 47

Text #16

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.5952. Min Act: -0.0000

Data Index: 345924 (C4 (Web Text))

Max Activating Token Index: 851

Text #17

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.4596. Min Act: -0.0000

Data Index: 279751 (C4 (Web Text))

Max Activating Token Index: 180

Text #18

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.5735. Min Act: -0.0000

Data Index: 1191131 (C4 (Web Text))

Max Activating Token Index: 18

Text #19

Max Range: 3.5086. Min Range: -3.5086

Max Act: 2.4281. Min Act: -0.0000

Data Index: 431001 (C4 (Web Text))

Max Activating Token Index: 776