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

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

Neuron 839 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.5003. Min Act: -0.0000

Data Index: 829969 (C4 (Web Text))

Max Activating Token Index: 983

Text #1

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.4890. Min Act: -0.0000

Data Index: 7440 (C4 (Web Text))

Max Activating Token Index: 401

Text #2

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.3200. Min Act: -0.0000

Data Index: 395270 (C4 (Web Text))

Max Activating Token Index: 216

Text #3

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.3402. Min Act: -0.0000

Data Index: 675901 (C4 (Web Text))

Max Activating Token Index: 15

Text #4

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2666. Min Act: -0.0000

Data Index: 572668 (C4 (Web Text))

Max Activating Token Index: 573

Text #5

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2714. Min Act: -0.0000

Data Index: 361705 (C4 (Web Text))

Max Activating Token Index: 339

Text #6

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.3124. Min Act: -0.0000

Data Index: 345079 (C4 (Web Text))

Max Activating Token Index: 517

Text #7

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2700. Min Act: -0.0000

Data Index: 382070 (C4 (Web Text))

Max Activating Token Index: 169

Text #8

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2643. Min Act: -0.0000

Data Index: 344150 (C4 (Web Text))

Max Activating Token Index: 627

Text #9

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2476. Min Act: -0.0000

Data Index: 1208485 (C4 (Web Text))

Max Activating Token Index: 858

Text #10

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2739. Min Act: -0.0000

Data Index: 643250 (C4 (Web Text))

Max Activating Token Index: 724

Text #11

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.1496. Min Act: -0.0000

Data Index: 1580789 (Python Code)

Max Activating Token Index: 639

Text #12

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2524. Min Act: -0.0000

Data Index: 49893 (C4 (Web Text))

Max Activating Token Index: 877

Text #13

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.1770. Min Act: -0.0000

Data Index: 1326177 (C4 (Web Text))

Max Activating Token Index: 152

Text #14

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.1975. Min Act: -0.0001

Data Index: 832122 (C4 (Web Text))

Max Activating Token Index: 966

Text #15

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2331. Min Act: -0.0000

Data Index: 1245069 (C4 (Web Text))

Max Activating Token Index: 634

Text #16

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2274. Min Act: -0.0000

Data Index: 966295 (C4 (Web Text))

Max Activating Token Index: 932

Text #17

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2722. Min Act: -0.0000

Data Index: 454678 (C4 (Web Text))

Max Activating Token Index: 21

Text #18

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.1891. Min Act: -0.0001

Data Index: 474947 (C4 (Web Text))

Max Activating Token Index: 488

Text #19

Max Range: 1.5003. Min Range: -1.5003

Max Act: 1.2229. Min Act: -0.0000

Data Index: 984863 (C4 (Web Text))

Max Activating Token Index: 569