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

Dataset: The Pile

Neuron 2420 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.0469. Min Range: -3.0469

Max Act: 3.0469. Min Act: -0.0000

Data Index: 726052 (The Pile)

Max Activating Token Index: 869

Text #1

Max Range: 3.0469. Min Range: -3.0469

Max Act: 2.7754. Min Act: -0.0000

Data Index: 1908350 (The Pile)

Max Activating Token Index: 176

Text #2

Max Range: 3.0469. Min Range: -3.0469

Max Act: 2.5555. Min Act: -0.0000

Data Index: 1767421 (The Pile)

Max Activating Token Index: 395

Text #3

Max Range: 3.0469. Min Range: -3.0469

Max Act: 2.7344. Min Act: -0.0000

Data Index: 700986 (The Pile)

Max Activating Token Index: 194

Text #4

Max Range: 3.0469. Min Range: -3.0469

Max Act: 2.1818. Min Act: -0.0000

Data Index: 1508308 (The Pile)

Max Activating Token Index: 733

Text #5

Max Range: 3.0469. Min Range: -3.0469

Max Act: 2.3914. Min Act: -0.0000

Data Index: 1404062 (The Pile)

Max Activating Token Index: 157

Text #6

Max Range: 3.0469. Min Range: -3.0469

Max Act: 2.0036. Min Act: -0.0000

Data Index: 37166 (The Pile)

Max Activating Token Index: 768

Text #7

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.8767. Min Act: -0.0000

Data Index: 590447 (The Pile)

Max Activating Token Index: 639

Text #8

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.9775. Min Act: -0.0000

Data Index: 1133482 (The Pile)

Max Activating Token Index: 107

Text #9

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.9326. Min Act: -0.0000

Data Index: 1047790 (The Pile)

Max Activating Token Index: 712

Text #10

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.9610. Min Act: -0.0000

Data Index: 1407205 (The Pile)

Max Activating Token Index: 570

Text #11

Max Range: 3.0469. Min Range: -3.0469

Max Act: 2.1906. Min Act: -0.0000

Data Index: 664416 (The Pile)

Max Activating Token Index: 547

Text #12

Max Range: 3.0469. Min Range: -3.0469

Max Act: 2.0329. Min Act: -0.0000

Data Index: 1506232 (The Pile)

Max Activating Token Index: 604

Text #13

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.8697. Min Act: -0.0000

Data Index: 1892186 (The Pile)

Max Activating Token Index: 317

Text #14

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.8904. Min Act: -0.0000

Data Index: 1260120 (The Pile)

Max Activating Token Index: 716

Text #15

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.8397. Min Act: -0.0000

Data Index: 660348 (The Pile)

Max Activating Token Index: 346

Text #16

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.9668. Min Act: -0.0000

Data Index: 562293 (The Pile)

Max Activating Token Index: 987

Text #17

Max Range: 3.0469. Min Range: -3.0469

Max Act: 2.0016. Min Act: -0.0000

Data Index: 676558 (The Pile)

Max Activating Token Index: 541

Text #18

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.9883. Min Act: -0.0000

Data Index: 1953336 (The Pile)

Max Activating Token Index: 577

Text #19

Max Range: 3.0469. Min Range: -3.0469

Max Act: 1.7515. Min Act: -0.0000

Data Index: 1268920 (The Pile)

Max Activating Token Index: 947