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

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

Neuron 981 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: 0.8740. Min Range: -0.8740

Max Act: 0.7683. Min Act: -0.0001

Data Index: 1180255 (C4 (Web Text))

Max Activating Token Index: 601

Text #1

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.7253. Min Act: -0.0001

Data Index: 946737 (C4 (Web Text))

Max Activating Token Index: 827

Text #2

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6887. Min Act: -0.0002

Data Index: 182698 (C4 (Web Text))

Max Activating Token Index: 236

Text #3

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.8740. Min Act: -0.0002

Data Index: 26365 (C4 (Web Text))

Max Activating Token Index: 631

Text #4

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.7846. Min Act: -0.0002

Data Index: 328133 (C4 (Web Text))

Max Activating Token Index: 440

Text #5

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.7588. Min Act: -0.0001

Data Index: 1276965 (C4 (Web Text))

Max Activating Token Index: 876

Text #6

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6687. Min Act: -0.0001

Data Index: 121001 (C4 (Web Text))

Max Activating Token Index: 640

Text #7

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6637. Min Act: -0.0002

Data Index: 1307715 (C4 (Web Text))

Max Activating Token Index: 267

Text #8

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6955. Min Act: -0.0001

Data Index: 1368172 (Python Code)

Max Activating Token Index: 274

Text #9

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6737. Min Act: -0.0002

Data Index: 228708 (C4 (Web Text))

Max Activating Token Index: 785

Text #10

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.7186. Min Act: -0.0001

Data Index: 1407986 (Python Code)

Max Activating Token Index: 594

Text #11

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6972. Min Act: -0.0002

Data Index: 865567 (C4 (Web Text))

Max Activating Token Index: 95

Text #12

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6598. Min Act: -0.0002

Data Index: 1008008 (C4 (Web Text))

Max Activating Token Index: 488

Text #13

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6676. Min Act: -0.0002

Data Index: 1099839 (C4 (Web Text))

Max Activating Token Index: 666

Text #14

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6847. Min Act: -0.0001

Data Index: 1631688 (Python Code)

Max Activating Token Index: 440

Text #15

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.7013. Min Act: -0.0002

Data Index: 965271 (C4 (Web Text))

Max Activating Token Index: 275

Text #16

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6414. Min Act: -0.0002

Data Index: 653832 (C4 (Web Text))

Max Activating Token Index: 155

Text #17

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6976. Min Act: -0.0001

Data Index: 1427143 (Python Code)

Max Activating Token Index: 45

Text #18

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.6062. Min Act: -0.0002

Data Index: 241016 (C4 (Web Text))

Max Activating Token Index: 564

Text #19

Max Range: 0.8740. Min Range: -0.8740

Max Act: 0.7013. Min Act: -0.0002

Data Index: 580616 (C4 (Web Text))

Max Activating Token Index: 553