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

Dataset: The Pile

Neuron 4385 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.2199. Min Range: -8.2199

Max Act: 8.2199. Min Act: -0.0000

Data Index: 8669 (The Pile)

Max Activating Token Index: 266

Text #1

Max Range: 8.2199. Min Range: -8.2199

Max Act: 7.5123. Min Act: -0.0000

Data Index: 382503 (The Pile)

Max Activating Token Index: 707

Text #2

Max Range: 8.2199. Min Range: -8.2199

Max Act: 6.9433. Min Act: -0.0000

Data Index: 1954936 (The Pile)

Max Activating Token Index: 900

Text #3

Max Range: 8.2199. Min Range: -8.2199

Max Act: 6.6830. Min Act: -0.0000

Data Index: 799329 (The Pile)

Max Activating Token Index: 88

Text #4

Max Range: 8.2199. Min Range: -8.2199

Max Act: 6.6420. Min Act: -0.0000

Data Index: 1414502 (The Pile)

Max Activating Token Index: 556

Text #5

Max Range: 8.2199. Min Range: -8.2199

Max Act: 6.7120. Min Act: -0.0000

Data Index: 60301 (The Pile)

Max Activating Token Index: 525

Text #6

Max Range: 8.2199. Min Range: -8.2199

Max Act: 6.3292. Min Act: -0.0000

Data Index: 1560306 (The Pile)

Max Activating Token Index: 693

Text #7

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.9102. Min Act: -0.0000

Data Index: 226700 (The Pile)

Max Activating Token Index: 18

Text #8

Max Range: 8.2199. Min Range: -8.2199

Max Act: 6.2325. Min Act: -0.0000

Data Index: 1091657 (The Pile)

Max Activating Token Index: 508

Text #9

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.9982. Min Act: -0.0000

Data Index: 1696854 (The Pile)

Max Activating Token Index: 445

Text #10

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.7913. Min Act: -0.0000

Data Index: 990130 (The Pile)

Max Activating Token Index: 283

Text #11

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.7842. Min Act: -0.0000

Data Index: 563019 (The Pile)

Max Activating Token Index: 825

Text #12

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.9026. Min Act: -0.0000

Data Index: 1095548 (The Pile)

Max Activating Token Index: 67

Text #13

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.7257. Min Act: -0.0000

Data Index: 311664 (The Pile)

Max Activating Token Index: 413

Text #14

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.4762. Min Act: -0.0000

Data Index: 1184231 (The Pile)

Max Activating Token Index: 247

Text #15

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.6354. Min Act: -0.0000

Data Index: 1397876 (The Pile)

Max Activating Token Index: 714

Text #16

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.4575. Min Act: -0.0000

Data Index: 1377537 (The Pile)

Max Activating Token Index: 476

Text #17

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.3863. Min Act: -0.0000

Data Index: 54331 (The Pile)

Max Activating Token Index: 244

Text #18

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.5600. Min Act: -0.0000

Data Index: 1597770 (The Pile)

Max Activating Token Index: 790

Text #19

Max Range: 8.2199. Min Range: -8.2199

Max Act: 5.5269. Min Act: -0.0000

Data Index: 1220960 (The Pile)

Max Activating Token Index: 405