Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 654 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.5089. Min Act: -0.0001

Data Index: 824127 (C4 (Web Text))

Max Activating Token Index: 76

Text #1

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.5115. Min Act: -0.0001

Data Index: 971411 (C4 (Web Text))

Max Activating Token Index: 380

Text #2

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4656. Min Act: -0.0001

Data Index: 86499 (C4 (Web Text))

Max Activating Token Index: 247

Text #3

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.5099. Min Act: -0.0001

Data Index: 1191415 (C4 (Web Text))

Max Activating Token Index: 57

Text #4

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.5420. Min Act: -0.0001

Data Index: 584549 (C4 (Web Text))

Max Activating Token Index: 884

Text #5

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.5507. Min Act: -0.0001

Data Index: 615990 (C4 (Web Text))

Max Activating Token Index: 787

Text #6

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4657. Min Act: -0.0001

Data Index: 844018 (C4 (Web Text))

Max Activating Token Index: 264

Text #7

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.5160. Min Act: -0.0001

Data Index: 130640 (C4 (Web Text))

Max Activating Token Index: 145

Text #8

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4571. Min Act: -0.0001

Data Index: 1095637 (C4 (Web Text))

Max Activating Token Index: 341

Text #9

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4903. Min Act: -0.0001

Data Index: 1350476 (C4 (Web Text))

Max Activating Token Index: 878

Text #10

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4776. Min Act: -0.0001

Data Index: 813806 (C4 (Web Text))

Max Activating Token Index: 800

Text #11

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.5023. Min Act: -0.0001

Data Index: 1039187 (C4 (Web Text))

Max Activating Token Index: 337

Text #12

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4828. Min Act: -0.0001

Data Index: 182055 (C4 (Web Text))

Max Activating Token Index: 249

Text #13

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4665. Min Act: -0.0001

Data Index: 1268568 (C4 (Web Text))

Max Activating Token Index: 170

Text #14

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4759. Min Act: -0.0001

Data Index: 601977 (C4 (Web Text))

Max Activating Token Index: 231

Text #15

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4641. Min Act: -0.0001

Data Index: 745278 (C4 (Web Text))

Max Activating Token Index: 249

Text #16

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4715. Min Act: -0.0001

Data Index: 236419 (C4 (Web Text))

Max Activating Token Index: 155

Text #17

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4551. Min Act: -0.0001

Data Index: 137062 (C4 (Web Text))

Max Activating Token Index: 299

Text #18

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4679. Min Act: -0.0001

Data Index: 1016719 (C4 (Web Text))

Max Activating Token Index: 465

Text #19

Max Range: 0.5507. Min Range: -0.5507

Max Act: 0.4546. Min Act: -0.0001

Data Index: 193259 (C4 (Web Text))

Max Activating Token Index: 70