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

Dataset: The Pile

Neuron 414 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: 8.3187. Min Range: -8.3187

Max Act: 8.3187. Min Act: -0.0000

Data Index: 343704 (The Pile)

Max Activating Token Index: 311

Text #1

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.3118. Min Act: -0.0000

Data Index: 1424951 (The Pile)

Max Activating Token Index: 38

Text #2

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.7702. Min Act: -0.0000

Data Index: 1867227 (The Pile)

Max Activating Token Index: 201

Text #3

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.6119. Min Act: -0.0000

Data Index: 115773 (The Pile)

Max Activating Token Index: 53

Text #4

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.3063. Min Act: -0.0000

Data Index: 1204086 (The Pile)

Max Activating Token Index: 325

Text #5

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.4322. Min Act: -0.0000

Data Index: 739821 (The Pile)

Max Activating Token Index: 414

Text #6

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.4428. Min Act: -0.0000

Data Index: 1639175 (The Pile)

Max Activating Token Index: 719

Text #7

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.1925. Min Act: -0.0000

Data Index: 1365131 (The Pile)

Max Activating Token Index: 105

Text #8

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.3101. Min Act: -0.0000

Data Index: 1098523 (The Pile)

Max Activating Token Index: 556

Text #9

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.2629. Min Act: -0.0000

Data Index: 1588494 (The Pile)

Max Activating Token Index: 185

Text #10

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.2387. Min Act: -0.0000

Data Index: 1837479 (The Pile)

Max Activating Token Index: 22

Text #11

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.2296. Min Act: -0.0000

Data Index: 1407811 (The Pile)

Max Activating Token Index: 707

Text #12

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.1778. Min Act: -0.0000

Data Index: 317442 (The Pile)

Max Activating Token Index: 478

Text #13

Max Range: 8.3187. Min Range: -8.3187

Max Act: 6.9943. Min Act: -0.0000

Data Index: 475107 (The Pile)

Max Activating Token Index: 207

Text #14

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.1693. Min Act: -0.0000

Data Index: 1303118 (The Pile)

Max Activating Token Index: 447

Text #15

Max Range: 8.3187. Min Range: -8.3187

Max Act: 6.9520. Min Act: -0.0000

Data Index: 1953322 (The Pile)

Max Activating Token Index: 429

Text #16

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.1285. Min Act: -0.0000

Data Index: 1014365 (The Pile)

Max Activating Token Index: 312

Text #17

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.2117. Min Act: -0.0000

Data Index: 381756 (The Pile)

Max Activating Token Index: 9

Text #18

Max Range: 8.3187. Min Range: -8.3187

Max Act: 7.1573. Min Act: -0.0000

Data Index: 1567918 (The Pile)

Max Activating Token Index: 598

Text #19

Max Range: 8.3187. Min Range: -8.3187

Max Act: 6.8044. Min Act: -0.0000

Data Index: 1282180 (The Pile)

Max Activating Token Index: 355