Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3154 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.4995. Min Act: -0.0000

Data Index: 48995 (C4 (Web Text))

Max Activating Token Index: 170

Text #1

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.4423. Min Act: -0.0000

Data Index: 645071 (C4 (Web Text))

Max Activating Token Index: 567

Text #2

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.4239. Min Act: -0.0000

Data Index: 701921 (C4 (Web Text))

Max Activating Token Index: 574

Text #3

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.4156. Min Act: -0.0000

Data Index: 832889 (C4 (Web Text))

Max Activating Token Index: 579

Text #4

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3854. Min Act: -0.0000

Data Index: 98981 (C4 (Web Text))

Max Activating Token Index: 540

Text #5

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.4359. Min Act: -0.0000

Data Index: 270497 (C4 (Web Text))

Max Activating Token Index: 989

Text #6

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.4454. Min Act: -0.0000

Data Index: 1198253 (C4 (Web Text))

Max Activating Token Index: 334

Text #7

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3676. Min Act: -0.0000

Data Index: 753421 (C4 (Web Text))

Max Activating Token Index: 323

Text #8

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3765. Min Act: -0.0000

Data Index: 1429117 (Python Code)

Max Activating Token Index: 294

Text #9

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3489. Min Act: -0.0000

Data Index: 1262341 (C4 (Web Text))

Max Activating Token Index: 773

Text #10

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3938. Min Act: -0.0000

Data Index: 357818 (C4 (Web Text))

Max Activating Token Index: 978

Text #11

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3324. Min Act: -0.0000

Data Index: 1604777 (Python Code)

Max Activating Token Index: 1005

Text #12

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3555. Min Act: -0.0000

Data Index: 10092 (C4 (Web Text))

Max Activating Token Index: 298

Text #13

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3198. Min Act: -0.0000

Data Index: 1354048 (C4 (Web Text))

Max Activating Token Index: 656

Text #14

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3260. Min Act: -0.0000

Data Index: 220597 (C4 (Web Text))

Max Activating Token Index: 23

Text #15

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3316. Min Act: -0.0000

Data Index: 1371143 (Python Code)

Max Activating Token Index: 838

Text #16

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3077. Min Act: -0.0000

Data Index: 1389276 (Python Code)

Max Activating Token Index: 720

Text #17

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.3112. Min Act: -0.0000

Data Index: 1402180 (Python Code)

Max Activating Token Index: 834

Text #18

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.2869. Min Act: -0.0000

Data Index: 1624009 (Python Code)

Max Activating Token Index: 667

Text #19

Max Range: 0.4995. Min Range: -0.4995

Max Act: 0.2503. Min Act: -0.0000

Data Index: 1417362 (Python Code)

Max Activating Token Index: 821