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

Dataset: The Pile

Neuron 825 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.0923. Min Range: -8.0923

Max Act: 8.0923. Min Act: -0.0000

Data Index: 1925699 (The Pile)

Max Activating Token Index: 221

Text #1

Max Range: 8.0923. Min Range: -8.0923

Max Act: 7.5499. Min Act: -0.0000

Data Index: 1530559 (The Pile)

Max Activating Token Index: 351

Text #2

Max Range: 8.0923. Min Range: -8.0923

Max Act: 7.3053. Min Act: -0.0000

Data Index: 1043689 (The Pile)

Max Activating Token Index: 148

Text #3

Max Range: 8.0923. Min Range: -8.0923

Max Act: 7.1015. Min Act: -0.0000

Data Index: 441392 (The Pile)

Max Activating Token Index: 112

Text #4

Max Range: 8.0923. Min Range: -8.0923

Max Act: 7.3460. Min Act: -0.0000

Data Index: 1534528 (The Pile)

Max Activating Token Index: 399

Text #5

Max Range: 8.0923. Min Range: -8.0923

Max Act: 7.1082. Min Act: -0.0000

Data Index: 365856 (The Pile)

Max Activating Token Index: 289

Text #6

Max Range: 8.0923. Min Range: -8.0923

Max Act: 6.9622. Min Act: -0.0000

Data Index: 877352 (The Pile)

Max Activating Token Index: 147

Text #7

Max Range: 8.0923. Min Range: -8.0923

Max Act: 6.5129. Min Act: -0.0000

Data Index: 914488 (The Pile)

Max Activating Token Index: 566

Text #8

Max Range: 8.0923. Min Range: -8.0923

Max Act: 6.5184. Min Act: -0.0000

Data Index: 632137 (The Pile)

Max Activating Token Index: 163

Text #9

Max Range: 8.0923. Min Range: -8.0923

Max Act: 6.2580. Min Act: -0.0000

Data Index: 319392 (The Pile)

Max Activating Token Index: 235

Text #10

Max Range: 8.0923. Min Range: -8.0923

Max Act: 6.1799. Min Act: -0.0000

Data Index: 842523 (The Pile)

Max Activating Token Index: 185

Text #11

Max Range: 8.0923. Min Range: -8.0923

Max Act: 6.2977. Min Act: -0.0000

Data Index: 22822 (The Pile)

Max Activating Token Index: 321

Text #12

Max Range: 8.0923. Min Range: -8.0923

Max Act: 6.0565. Min Act: -0.0000

Data Index: 1546708 (The Pile)

Max Activating Token Index: 220

Text #13

Max Range: 8.0923. Min Range: -8.0923

Max Act: 6.2416. Min Act: -0.0000

Data Index: 830263 (The Pile)

Max Activating Token Index: 224

Text #14

Max Range: 8.0923. Min Range: -8.0923

Max Act: 5.7748. Min Act: -0.0000

Data Index: 1506595 (The Pile)

Max Activating Token Index: 124

Text #15

Max Range: 8.0923. Min Range: -8.0923

Max Act: 5.9531. Min Act: -0.0000

Data Index: 314450 (The Pile)

Max Activating Token Index: 194

Text #16

Max Range: 8.0923. Min Range: -8.0923

Max Act: 6.0464. Min Act: -0.0000

Data Index: 167639 (The Pile)

Max Activating Token Index: 313

Text #17

Max Range: 8.0923. Min Range: -8.0923

Max Act: 5.6334. Min Act: -0.0000

Data Index: 1575626 (The Pile)

Max Activating Token Index: 192

Text #18

Max Range: 8.0923. Min Range: -8.0923

Max Act: 5.6040. Min Act: -0.0000

Data Index: 250503 (The Pile)

Max Activating Token Index: 798

Text #19

Max Range: 8.0923. Min Range: -8.0923

Max Act: 5.2744. Min Act: -0.0000

Data Index: 167593 (The Pile)

Max Activating Token Index: 711