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

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

Neuron 3865 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.8022. Min Act: -0.0001

Data Index: 1472067 (Python Code)

Max Activating Token Index: 150

Text #1

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.3183. Min Act: -0.0000

Data Index: 535799 (C4 (Web Text))

Max Activating Token Index: 167

Text #2

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.2600. Min Act: -0.0000

Data Index: 154630 (C4 (Web Text))

Max Activating Token Index: 668

Text #3

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.1355. Min Act: -0.0001

Data Index: 357238 (C4 (Web Text))

Max Activating Token Index: 382

Text #4

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.0839. Min Act: -0.0001

Data Index: 553273 (C4 (Web Text))

Max Activating Token Index: 327

Text #5

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.0954. Min Act: -0.0000

Data Index: 86439 (C4 (Web Text))

Max Activating Token Index: 465

Text #6

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.1851. Min Act: -0.0001

Data Index: 951079 (C4 (Web Text))

Max Activating Token Index: 690

Text #7

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.0404. Min Act: -0.0001

Data Index: 1438028 (Python Code)

Max Activating Token Index: 232

Text #8

Max Range: 1.8022. Min Range: -1.8022

Max Act: 0.9899. Min Act: -0.0001

Data Index: 128913 (C4 (Web Text))

Max Activating Token Index: 598

Text #9

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.0358. Min Act: -0.0001

Data Index: 212940 (C4 (Web Text))

Max Activating Token Index: 770

Text #10

Max Range: 1.8022. Min Range: -1.8022

Max Act: 0.9937. Min Act: -0.0001

Data Index: 1167493 (C4 (Web Text))

Max Activating Token Index: 92

Text #11

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.0518. Min Act: -0.0001

Data Index: 415884 (C4 (Web Text))

Max Activating Token Index: 399

Text #12

Max Range: 1.8022. Min Range: -1.8022

Max Act: 0.9997. Min Act: -0.0001

Data Index: 22743 (C4 (Web Text))

Max Activating Token Index: 677

Text #13

Max Range: 1.8022. Min Range: -1.8022

Max Act: 1.0192. Min Act: -0.0000

Data Index: 488354 (C4 (Web Text))

Max Activating Token Index: 731

Text #14

Max Range: 1.8022. Min Range: -1.8022

Max Act: 0.9429. Min Act: -0.0001

Data Index: 136513 (C4 (Web Text))

Max Activating Token Index: 959

Text #15

Max Range: 1.8022. Min Range: -1.8022

Max Act: 0.9570. Min Act: -0.0001

Data Index: 312780 (C4 (Web Text))

Max Activating Token Index: 635

Text #16

Max Range: 1.8022. Min Range: -1.8022

Max Act: 0.8956. Min Act: -0.0001

Data Index: 1260909 (C4 (Web Text))

Max Activating Token Index: 701

Text #17

Max Range: 1.8022. Min Range: -1.8022

Max Act: 0.9440. Min Act: -0.0001

Data Index: 1147653 (C4 (Web Text))

Max Activating Token Index: 216

Text #18

Max Range: 1.8022. Min Range: -1.8022

Max Act: 0.8784. Min Act: -0.0001

Data Index: 36065 (C4 (Web Text))

Max Activating Token Index: 460

Text #19

Max Range: 1.8022. Min Range: -1.8022

Max Act: 0.8634. Min Act: -0.0001

Data Index: 845998 (C4 (Web Text))

Max Activating Token Index: 755