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

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

Neuron 1604 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.7008. Min Act: -0.0000

Data Index: 278334 (C4 (Web Text))

Max Activating Token Index: 897

Text #1

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.6295. Min Act: -0.0001

Data Index: 720467 (C4 (Web Text))

Max Activating Token Index: 347

Text #2

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.6287. Min Act: -0.0000

Data Index: 271046 (C4 (Web Text))

Max Activating Token Index: 183

Text #3

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.5041. Min Act: -0.0000

Data Index: 1492361 (Python Code)

Max Activating Token Index: 58

Text #4

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.4718. Min Act: -0.0000

Data Index: 957823 (C4 (Web Text))

Max Activating Token Index: 508

Text #5

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.4711. Min Act: -0.0000

Data Index: 299002 (C4 (Web Text))

Max Activating Token Index: 16

Text #6

Max Range: 2.7008. Min Range: -2.7008

Max Act: 1.8239. Min Act: -0.0001

Data Index: 348097 (C4 (Web Text))

Max Activating Token Index: 309

Text #7

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.3075. Min Act: -0.0000

Data Index: 5830 (C4 (Web Text))

Max Activating Token Index: 794

Text #8

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.3080. Min Act: -0.0001

Data Index: 18392 (C4 (Web Text))

Max Activating Token Index: 362

Text #9

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.0597. Min Act: -0.0001

Data Index: 1026190 (C4 (Web Text))

Max Activating Token Index: 116

Text #10

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.4015. Min Act: -0.0000

Data Index: 1136441 (C4 (Web Text))

Max Activating Token Index: 829

Text #11

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.2289. Min Act: -0.0000

Data Index: 195005 (C4 (Web Text))

Max Activating Token Index: 242

Text #12

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.2174. Min Act: -0.0000

Data Index: 676308 (C4 (Web Text))

Max Activating Token Index: 909

Text #13

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.2059. Min Act: -0.0000

Data Index: 1099565 (C4 (Web Text))

Max Activating Token Index: 756

Text #14

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.1531. Min Act: -0.0000

Data Index: 733692 (C4 (Web Text))

Max Activating Token Index: 615

Text #15

Max Range: 2.7008. Min Range: -2.7008

Max Act: 1.8508. Min Act: -0.0001

Data Index: 125569 (C4 (Web Text))

Max Activating Token Index: 721

Text #16

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.0047. Min Act: -0.0001

Data Index: 692371 (C4 (Web Text))

Max Activating Token Index: 589

Text #17

Max Range: 2.7008. Min Range: -2.7008

Max Act: 1.8978. Min Act: -0.0000

Data Index: 1191030 (C4 (Web Text))

Max Activating Token Index: 757

Text #18

Max Range: 2.7008. Min Range: -2.7008

Max Act: 1.9479. Min Act: -0.0000

Data Index: 399383 (C4 (Web Text))

Max Activating Token Index: 264

Text #19

Max Range: 2.7008. Min Range: -2.7008

Max Act: 2.0457. Min Act: -0.0000

Data Index: 102591 (C4 (Web Text))

Max Activating Token Index: 214