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

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

Neuron 433 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.8162. Min Act: -0.0001

Data Index: 428165 (C4 (Web Text))

Max Activating Token Index: 850

Text #1

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.7752. Min Act: -0.0001

Data Index: 170701 (C4 (Web Text))

Max Activating Token Index: 365

Text #2

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.5385. Min Act: -0.0001

Data Index: 382180 (C4 (Web Text))

Max Activating Token Index: 892

Text #3

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.3326. Min Act: -0.0000

Data Index: 236652 (C4 (Web Text))

Max Activating Token Index: 899

Text #4

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.1655. Min Act: -0.0000

Data Index: 773592 (C4 (Web Text))

Max Activating Token Index: 814

Text #5

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.3265. Min Act: -0.0000

Data Index: 815439 (C4 (Web Text))

Max Activating Token Index: 446

Text #6

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.2545. Min Act: -0.0000

Data Index: 272996 (C4 (Web Text))

Max Activating Token Index: 899

Text #7

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.2390. Min Act: -0.0000

Data Index: 642512 (C4 (Web Text))

Max Activating Token Index: 557

Text #8

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.2478. Min Act: -0.0000

Data Index: 907659 (C4 (Web Text))

Max Activating Token Index: 771

Text #9

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.6156. Min Act: -0.0000

Data Index: 529548 (C4 (Web Text))

Max Activating Token Index: 652

Text #10

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.2457. Min Act: -0.0000

Data Index: 975008 (C4 (Web Text))

Max Activating Token Index: 490

Text #11

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.0492. Min Act: -0.0000

Data Index: 266012 (C4 (Web Text))

Max Activating Token Index: 605

Text #12

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.2332. Min Act: -0.0000

Data Index: 102744 (C4 (Web Text))

Max Activating Token Index: 536

Text #13

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.2528. Min Act: -0.0001

Data Index: 1612976 (Python Code)

Max Activating Token Index: 664

Text #14

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.1673. Min Act: -0.0001

Data Index: 1065931 (C4 (Web Text))

Max Activating Token Index: 502

Text #15

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.0524. Min Act: -0.0000

Data Index: 150005 (C4 (Web Text))

Max Activating Token Index: 430

Text #16

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.0440. Min Act: -0.0001

Data Index: 1551439 (Python Code)

Max Activating Token Index: 429

Text #17

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.1236. Min Act: -0.0001

Data Index: 740905 (C4 (Web Text))

Max Activating Token Index: 567

Text #18

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.1503. Min Act: -0.0000

Data Index: 668486 (C4 (Web Text))

Max Activating Token Index: 25

Text #19

Max Range: 2.8162. Min Range: -2.8162

Max Act: 2.0252. Min Act: -0.0001

Data Index: 1618232 (Python Code)

Max Activating Token Index: 995