Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2074 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0143. Min Act: -0.0001

Data Index: 280054 (C4 (Web Text))

Max Activating Token Index: 328

Text #1

Max Range: 1.4653. Min Range: -1.4653

Max Act: 1.4653. Min Act: -0.0001

Data Index: 1004146 (C4 (Web Text))

Max Activating Token Index: 314

Text #2

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.2240. Min Act: -0.0001

Data Index: 422549 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.1683. Min Act: -0.0001

Data Index: 1581359 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.2259. Min Act: -0.0001

Data Index: 537726 (C4 (Web Text))

Max Activating Token Index: 242

Text #5

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0227. Min Act: -0.0001

Data Index: 1479789 (Python Code)

Max Activating Token Index: 297

Text #6

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0422. Min Act: -0.0001

Data Index: 323903 (C4 (Web Text))

Max Activating Token Index: 957

Text #7

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0823. Min Act: -0.0001

Data Index: 1195869 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0647. Min Act: -0.0001

Data Index: 214850 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0376. Min Act: -0.0001

Data Index: 1653860 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0017. Min Act: -0.0001

Data Index: 447345 (C4 (Web Text))

Max Activating Token Index: 859

Text #11

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0596. Min Act: -0.0001

Data Index: 1375101 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0040. Min Act: -0.0001

Data Index: 34710 (C4 (Web Text))

Max Activating Token Index: 266

Text #13

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0101. Min Act: -0.0001

Data Index: 437024 (C4 (Web Text))

Max Activating Token Index: 390

Text #14

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0119. Min Act: -0.0001

Data Index: 1341808 (C4 (Web Text))

Max Activating Token Index: 838

Text #15

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0801. Min Act: -0.0001

Data Index: 1019055 (C4 (Web Text))

Max Activating Token Index: 926

Text #16

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0374. Min Act: -0.0001

Data Index: 98164 (C4 (Web Text))

Max Activating Token Index: 580

Text #17

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0365. Min Act: -0.0001

Data Index: 827315 (C4 (Web Text))

Max Activating Token Index: 90

Text #18

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0067. Min Act: -0.0001

Data Index: 528277 (C4 (Web Text))

Max Activating Token Index: 368

Text #19

Max Range: 1.4653. Min Range: -1.4653

Max Act: 0.0337. Min Act: -0.0001

Data Index: 513245 (C4 (Web Text))

Max Activating Token Index: 993