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

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

Neuron 252 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.7778. Min Act: -0.0002

Data Index: 844316 (C4 (Web Text))

Max Activating Token Index: 17

Text #1

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.6247. Min Act: -0.0002

Data Index: 464670 (C4 (Web Text))

Max Activating Token Index: 954

Text #2

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.6025. Min Act: -0.0002

Data Index: 1102636 (C4 (Web Text))

Max Activating Token Index: 1010

Text #3

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5890. Min Act: -0.0002

Data Index: 629223 (C4 (Web Text))

Max Activating Token Index: 3

Text #4

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5696. Min Act: -0.0001

Data Index: 101239 (C4 (Web Text))

Max Activating Token Index: 53

Text #5

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5420. Min Act: -0.0001

Data Index: 1159477 (C4 (Web Text))

Max Activating Token Index: 673

Text #6

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5704. Min Act: -0.0001

Data Index: 758769 (C4 (Web Text))

Max Activating Token Index: 1005

Text #7

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5591. Min Act: -0.0001

Data Index: 1532 (C4 (Web Text))

Max Activating Token Index: 655

Text #8

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5612. Min Act: -0.0001

Data Index: 740204 (C4 (Web Text))

Max Activating Token Index: 147

Text #9

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5894. Min Act: -0.0001

Data Index: 1039961 (C4 (Web Text))

Max Activating Token Index: 986

Text #10

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5383. Min Act: -0.0001

Data Index: 643829 (C4 (Web Text))

Max Activating Token Index: 1020

Text #11

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5332. Min Act: -0.0001

Data Index: 1461158 (Python Code)

Max Activating Token Index: 454

Text #12

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5473. Min Act: -0.0001

Data Index: 103252 (C4 (Web Text))

Max Activating Token Index: 397

Text #13

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5478. Min Act: -0.0002

Data Index: 416756 (C4 (Web Text))

Max Activating Token Index: 790

Text #14

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5794. Min Act: -0.0002

Data Index: 483003 (C4 (Web Text))

Max Activating Token Index: 940

Text #15

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5417. Min Act: -0.0001

Data Index: 1470508 (Python Code)

Max Activating Token Index: 1013

Text #16

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5391. Min Act: -0.0002

Data Index: 296048 (C4 (Web Text))

Max Activating Token Index: 841

Text #17

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5328. Min Act: -0.0001

Data Index: 644838 (C4 (Web Text))

Max Activating Token Index: 362

Text #18

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.5022. Min Act: -0.0002

Data Index: 710506 (C4 (Web Text))

Max Activating Token Index: 816

Text #19

Max Range: 0.7778. Min Range: -0.7778

Max Act: 0.4808. Min Act: -0.0001

Data Index: 30164 (C4 (Web Text))

Max Activating Token Index: 955