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

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

Neuron 4850 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.7093. Min Act: -0.0003

Data Index: 416932 (C4 (Web Text))

Max Activating Token Index: 19

Text #1

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.6698. Min Act: -0.0003

Data Index: 976372 (C4 (Web Text))

Max Activating Token Index: 202

Text #2

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.6775. Min Act: -0.0003

Data Index: 116183 (C4 (Web Text))

Max Activating Token Index: 209

Text #3

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.6270. Min Act: -0.0003

Data Index: 33458 (C4 (Web Text))

Max Activating Token Index: 30

Text #4

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.5619. Min Act: -0.0003

Data Index: 760192 (C4 (Web Text))

Max Activating Token Index: 587

Text #5

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.5658. Min Act: -0.0003

Data Index: 891602 (C4 (Web Text))

Max Activating Token Index: 706

Text #6

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.5681. Min Act: -0.0003

Data Index: 591702 (C4 (Web Text))

Max Activating Token Index: 336

Text #7

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.5778. Min Act: -0.0003

Data Index: 966805 (C4 (Web Text))

Max Activating Token Index: 266

Text #8

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.5530. Min Act: -0.0002

Data Index: 272751 (C4 (Web Text))

Max Activating Token Index: 164

Text #9

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.5070. Min Act: -0.0003

Data Index: 1034590 (C4 (Web Text))

Max Activating Token Index: 117

Text #10

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.5184. Min Act: -0.0002

Data Index: 173865 (C4 (Web Text))

Max Activating Token Index: 70

Text #11

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.4850. Min Act: -0.0003

Data Index: 499081 (C4 (Web Text))

Max Activating Token Index: 198

Text #12

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.5017. Min Act: -0.0003

Data Index: 561750 (C4 (Web Text))

Max Activating Token Index: 97

Text #13

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.4392. Min Act: -0.0003

Data Index: 676907 (C4 (Web Text))

Max Activating Token Index: 543

Text #14

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.4491. Min Act: -0.0003

Data Index: 685656 (C4 (Web Text))

Max Activating Token Index: 274

Text #15

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.4850. Min Act: -0.0003

Data Index: 1129783 (C4 (Web Text))

Max Activating Token Index: 340

Text #16

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.4702. Min Act: -0.0003

Data Index: 352681 (C4 (Web Text))

Max Activating Token Index: 275

Text #17

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.4988. Min Act: -0.0003

Data Index: 563352 (C4 (Web Text))

Max Activating Token Index: 218

Text #18

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.4959. Min Act: -0.0003

Data Index: 1000559 (C4 (Web Text))

Max Activating Token Index: 365

Text #19

Max Range: 0.7093. Min Range: -0.7093

Max Act: 0.4779. Min Act: -0.0003

Data Index: 1020797 (C4 (Web Text))

Max Activating Token Index: 412