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

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

Neuron 1685 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.4163. Min Act: -0.0001

Data Index: 504464 (C4 (Web Text))

Max Activating Token Index: 394

Text #1

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.4287. Min Act: -0.0000

Data Index: 1640500 (Python Code)

Max Activating Token Index: 488

Text #2

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3784. Min Act: -0.0001

Data Index: 1590171 (Python Code)

Max Activating Token Index: 545

Text #3

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3744. Min Act: -0.0001

Data Index: 469410 (C4 (Web Text))

Max Activating Token Index: 499

Text #4

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3756. Min Act: -0.0001

Data Index: 1375548 (Python Code)

Max Activating Token Index: 120

Text #5

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3187. Min Act: -0.0001

Data Index: 1210404 (C4 (Web Text))

Max Activating Token Index: 521

Text #6

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3282. Min Act: -0.0001

Data Index: 1619440 (Python Code)

Max Activating Token Index: 758

Text #7

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3512. Min Act: -0.0000

Data Index: 1010693 (C4 (Web Text))

Max Activating Token Index: 430

Text #8

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3184. Min Act: -0.0000

Data Index: 687115 (C4 (Web Text))

Max Activating Token Index: 553

Text #9

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3381. Min Act: -0.0000

Data Index: 707225 (C4 (Web Text))

Max Activating Token Index: 456

Text #10

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3380. Min Act: -0.0001

Data Index: 1359862 (Python Code)

Max Activating Token Index: 923

Text #11

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3453. Min Act: -0.0001

Data Index: 869122 (C4 (Web Text))

Max Activating Token Index: 171

Text #12

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3100. Min Act: -0.0001

Data Index: 638308 (C4 (Web Text))

Max Activating Token Index: 719

Text #13

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3341. Min Act: -0.0001

Data Index: 132818 (C4 (Web Text))

Max Activating Token Index: 325

Text #14

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3123. Min Act: -0.0001

Data Index: 1188505 (C4 (Web Text))

Max Activating Token Index: 256

Text #15

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3108. Min Act: -0.0001

Data Index: 1531050 (Python Code)

Max Activating Token Index: 350

Text #16

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3082. Min Act: -0.0001

Data Index: 604882 (C4 (Web Text))

Max Activating Token Index: 86

Text #17

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.3164. Min Act: -0.0001

Data Index: 561196 (C4 (Web Text))

Max Activating Token Index: 421

Text #18

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.2971. Min Act: -0.0001

Data Index: 269458 (C4 (Web Text))

Max Activating Token Index: 636

Text #19

Max Range: 0.4287. Min Range: -0.4287

Max Act: 0.2938. Min Act: -0.0000

Data Index: 581007 (C4 (Web Text))

Max Activating Token Index: 91