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

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

Neuron 5766 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.3209. Min Act: -0.0000

Data Index: 692345 (C4 (Web Text))

Max Activating Token Index: 749

Text #1

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.3750. Min Act: -0.0000

Data Index: 395997 (C4 (Web Text))

Max Activating Token Index: 22

Text #2

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.3490. Min Act: -0.0000

Data Index: 904611 (C4 (Web Text))

Max Activating Token Index: 998

Text #3

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.2421. Min Act: -0.0000

Data Index: 368228 (C4 (Web Text))

Max Activating Token Index: 1021

Text #4

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1526. Min Act: -0.0000

Data Index: 1614626 (Python Code)

Max Activating Token Index: 853

Text #5

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1591. Min Act: -0.0000

Data Index: 384556 (C4 (Web Text))

Max Activating Token Index: 11

Text #6

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1538. Min Act: -0.0000

Data Index: 364900 (C4 (Web Text))

Max Activating Token Index: 987

Text #7

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1442. Min Act: -0.0000

Data Index: 1051002 (C4 (Web Text))

Max Activating Token Index: 747

Text #8

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1721. Min Act: -0.0000

Data Index: 1343421 (C4 (Web Text))

Max Activating Token Index: 963

Text #9

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1624. Min Act: -0.0000

Data Index: 159683 (C4 (Web Text))

Max Activating Token Index: 316

Text #10

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1162. Min Act: -0.0000

Data Index: 975832 (C4 (Web Text))

Max Activating Token Index: 533

Text #11

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1398. Min Act: -0.0000

Data Index: 518573 (C4 (Web Text))

Max Activating Token Index: 621

Text #12

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1868. Min Act: -0.0000

Data Index: 267792 (C4 (Web Text))

Max Activating Token Index: 222

Text #13

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1745. Min Act: -0.0000

Data Index: 899603 (C4 (Web Text))

Max Activating Token Index: 61

Text #14

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1292. Min Act: -0.0000

Data Index: 1387858 (Python Code)

Max Activating Token Index: 981

Text #15

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1236. Min Act: -0.0000

Data Index: 119203 (C4 (Web Text))

Max Activating Token Index: 982

Text #16

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1429. Min Act: -0.0000

Data Index: 321132 (C4 (Web Text))

Max Activating Token Index: 638

Text #17

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.1084. Min Act: -0.0000

Data Index: 1022251 (C4 (Web Text))

Max Activating Token Index: 466

Text #18

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.0398. Min Act: -0.0000

Data Index: 800115 (C4 (Web Text))

Max Activating Token Index: 417

Text #19

Max Range: 1.3750. Min Range: -1.3750

Max Act: 1.0454. Min Act: -0.0000

Data Index: 1550150 (Python Code)

Max Activating Token Index: 294