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

Dataset: The Pile

Neuron 2706 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: 6.5420. Min Range: -6.5420

Max Act: 6.5420. Min Act: -0.0000

Data Index: 1241049 (The Pile)

Max Activating Token Index: 537

Text #1

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.8567. Min Act: -0.0000

Data Index: 1254805 (The Pile)

Max Activating Token Index: 534

Text #2

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.7635. Min Act: -0.0000

Data Index: 1519741 (The Pile)

Max Activating Token Index: 296

Text #3

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.6687. Min Act: -0.0000

Data Index: 437276 (The Pile)

Max Activating Token Index: 303

Text #4

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.8406. Min Act: -0.0000

Data Index: 1614222 (The Pile)

Max Activating Token Index: 191

Text #5

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.7687. Min Act: -0.0000

Data Index: 1798985 (The Pile)

Max Activating Token Index: 140

Text #6

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.5145. Min Act: -0.0000

Data Index: 819146 (The Pile)

Max Activating Token Index: 457

Text #7

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.4061. Min Act: -0.0000

Data Index: 1560190 (The Pile)

Max Activating Token Index: 256

Text #8

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.2186. Min Act: -0.0000

Data Index: 1321683 (The Pile)

Max Activating Token Index: 692

Text #9

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.3102. Min Act: -0.0000

Data Index: 1712597 (The Pile)

Max Activating Token Index: 200

Text #10

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.6126. Min Act: -0.0000

Data Index: 271801 (The Pile)

Max Activating Token Index: 79

Text #11

Max Range: 6.5420. Min Range: -6.5420

Max Act: 4.8787. Min Act: -0.0000

Data Index: 617683 (The Pile)

Max Activating Token Index: 247

Text #12

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.4009. Min Act: -0.0000

Data Index: 1217619 (The Pile)

Max Activating Token Index: 289

Text #13

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.5163. Min Act: -0.0000

Data Index: 408856 (The Pile)

Max Activating Token Index: 679

Text #14

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.6878. Min Act: -0.0000

Data Index: 1653446 (The Pile)

Max Activating Token Index: 195

Text #15

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.2776. Min Act: -0.0000

Data Index: 461506 (The Pile)

Max Activating Token Index: 687

Text #16

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.2513. Min Act: -0.0000

Data Index: 24321 (The Pile)

Max Activating Token Index: 534

Text #17

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.0720. Min Act: -0.0000

Data Index: 1351206 (The Pile)

Max Activating Token Index: 500

Text #18

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.3468. Min Act: -0.0000

Data Index: 1546128 (The Pile)

Max Activating Token Index: 233

Text #19

Max Range: 6.5420. Min Range: -6.5420

Max Act: 5.7392. Min Act: -0.0000

Data Index: 1747544 (The Pile)

Max Activating Token Index: 335