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

Dataset: The Pile

Neuron 404 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.8685. Min Act: -0.0000

Data Index: 1359735 (The Pile)

Max Activating Token Index: 363

Text #1

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.4943. Min Act: -0.0000

Data Index: 610593 (The Pile)

Max Activating Token Index: 775

Text #2

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.0308. Min Act: -0.0000

Data Index: 780307 (The Pile)

Max Activating Token Index: 86

Text #3

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.0450. Min Act: -0.0000

Data Index: 417855 (The Pile)

Max Activating Token Index: 249

Text #4

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.2084. Min Act: -0.0000

Data Index: 1115563 (The Pile)

Max Activating Token Index: 226

Text #5

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.3578. Min Act: -0.0000

Data Index: 1249894 (The Pile)

Max Activating Token Index: 504

Text #6

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.2981. Min Act: -0.0000

Data Index: 567601 (The Pile)

Max Activating Token Index: 518

Text #7

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.3682. Min Act: -0.0000

Data Index: 136494 (The Pile)

Max Activating Token Index: 383

Text #8

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.1400. Min Act: -0.0000

Data Index: 593867 (The Pile)

Max Activating Token Index: 624

Text #9

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.0567. Min Act: -0.0000

Data Index: 574843 (The Pile)

Max Activating Token Index: 144

Text #10

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.2074. Min Act: -0.0000

Data Index: 194228 (The Pile)

Max Activating Token Index: 501

Text #11

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.0118. Min Act: -0.0000

Data Index: 265710 (The Pile)

Max Activating Token Index: 775

Text #12

Max Range: 7.8685. Min Range: -7.8685

Max Act: 6.9768. Min Act: -0.0000

Data Index: 1874147 (The Pile)

Max Activating Token Index: 257

Text #13

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.3564. Min Act: -0.0000

Data Index: 556306 (The Pile)

Max Activating Token Index: 787

Text #14

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.0623. Min Act: -0.0000

Data Index: 1403440 (The Pile)

Max Activating Token Index: 445

Text #15

Max Range: 7.8685. Min Range: -7.8685

Max Act: 6.7235. Min Act: -0.0000

Data Index: 1022167 (The Pile)

Max Activating Token Index: 974

Text #16

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.3733. Min Act: -0.0000

Data Index: 457742 (The Pile)

Max Activating Token Index: 849

Text #17

Max Range: 7.8685. Min Range: -7.8685

Max Act: 7.0953. Min Act: -0.0000

Data Index: 1263269 (The Pile)

Max Activating Token Index: 773

Text #18

Max Range: 7.8685. Min Range: -7.8685

Max Act: 6.8927. Min Act: -0.0000

Data Index: 1724869 (The Pile)

Max Activating Token Index: 740

Text #19

Max Range: 7.8685. Min Range: -7.8685

Max Act: 6.4289. Min Act: -0.0000

Data Index: 1186329 (The Pile)

Max Activating Token Index: 536