Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: The Pile

Neuron 648 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 3.8285. Min Range: -3.8285

Max Act: 3.6593. Min Act: -0.0000

Data Index: 348552 (The Pile)

Max Activating Token Index: 474

Text #1

Max Range: 3.8285. Min Range: -3.8285

Max Act: 3.8285. Min Act: -0.0000

Data Index: 803257 (The Pile)

Max Activating Token Index: 223

Text #2

Max Range: 3.8285. Min Range: -3.8285

Max Act: 3.7942. Min Act: -0.0000

Data Index: 1067712 (The Pile)

Max Activating Token Index: 140

Text #3

Max Range: 3.8285. Min Range: -3.8285

Max Act: 3.7288. Min Act: -0.0000

Data Index: 2756 (The Pile)

Max Activating Token Index: 349

Text #4

Max Range: 3.8285. Min Range: -3.8285

Max Act: 3.1453. Min Act: -0.0000

Data Index: 1079256 (The Pile)

Max Activating Token Index: 770

Text #5

Max Range: 3.8285. Min Range: -3.8285

Max Act: 3.4870. Min Act: -0.0000

Data Index: 1139727 (The Pile)

Max Activating Token Index: 438

Text #6

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.8726. Min Act: -0.0000

Data Index: 518772 (The Pile)

Max Activating Token Index: 425

Text #7

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.6927. Min Act: -0.0000

Data Index: 902491 (The Pile)

Max Activating Token Index: 240

Text #8

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.5190. Min Act: -0.0000

Data Index: 640279 (The Pile)

Max Activating Token Index: 353

Text #9

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.5286. Min Act: -0.0000

Data Index: 903410 (The Pile)

Max Activating Token Index: 451

Text #10

Max Range: 3.8285. Min Range: -3.8285

Max Act: 1.9290. Min Act: -0.0000

Data Index: 232117 (The Pile)

Max Activating Token Index: 368

Text #11

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.2015. Min Act: -0.0000

Data Index: 122875 (The Pile)

Max Activating Token Index: 848

Text #12

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.9233. Min Act: -0.0000

Data Index: 1639087 (The Pile)

Max Activating Token Index: 809

Text #13

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.2137. Min Act: -0.0000

Data Index: 963911 (The Pile)

Max Activating Token Index: 391

Text #14

Max Range: 3.8285. Min Range: -3.8285

Max Act: 1.9766. Min Act: -0.0000

Data Index: 1947868 (The Pile)

Max Activating Token Index: 327

Text #15

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.3452. Min Act: -0.0000

Data Index: 1106232 (The Pile)

Max Activating Token Index: 928

Text #16

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.0479. Min Act: -0.0000

Data Index: 1429865 (The Pile)

Max Activating Token Index: 898

Text #17

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.1555. Min Act: -0.0000

Data Index: 1750029 (The Pile)

Max Activating Token Index: 302

Text #18

Max Range: 3.8285. Min Range: -3.8285

Max Act: 2.0298. Min Act: -0.0000

Data Index: 436801 (The Pile)

Max Activating Token Index: 149

Text #19

Max Range: 3.8285. Min Range: -3.8285

Max Act: 1.7917. Min Act: -0.0000

Data Index: 1402161 (The Pile)

Max Activating Token Index: 720