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

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

Neuron 584 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.0358. Min Range: -2.0358

Max Act: 2.0358. Min Act: -0.0000

Data Index: 771556 (C4 (Web Text))

Max Activating Token Index: 837

Text #1

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.9249. Min Act: -0.0000

Data Index: 725998 (C4 (Web Text))

Max Activating Token Index: 543

Text #2

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.7872. Min Act: -0.0000

Data Index: 351417 (C4 (Web Text))

Max Activating Token Index: 338

Text #3

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.6511. Min Act: -0.0000

Data Index: 1190508 (C4 (Web Text))

Max Activating Token Index: 857

Text #4

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.5768. Min Act: -0.0000

Data Index: 45448 (C4 (Web Text))

Max Activating Token Index: 590

Text #5

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.4826. Min Act: -0.0000

Data Index: 1587810 (Python Code)

Max Activating Token Index: 159

Text #6

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.3940. Min Act: -0.0000

Data Index: 824774 (C4 (Web Text))

Max Activating Token Index: 325

Text #7

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.3883. Min Act: -0.0000

Data Index: 1078455 (C4 (Web Text))

Max Activating Token Index: 336

Text #8

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.3594. Min Act: -0.0000

Data Index: 1499604 (Python Code)

Max Activating Token Index: 667

Text #9

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.3315. Min Act: -0.0000

Data Index: 928029 (C4 (Web Text))

Max Activating Token Index: 835

Text #10

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.3229. Min Act: -0.0000

Data Index: 1585390 (Python Code)

Max Activating Token Index: 530

Text #11

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.3442. Min Act: -0.0000

Data Index: 1427617 (Python Code)

Max Activating Token Index: 109

Text #12

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.3665. Min Act: -0.0000

Data Index: 375642 (C4 (Web Text))

Max Activating Token Index: 671

Text #13

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.3367. Min Act: -0.0000

Data Index: 1556819 (Python Code)

Max Activating Token Index: 82

Text #14

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.3849. Min Act: -0.0000

Data Index: 1259749 (C4 (Web Text))

Max Activating Token Index: 266

Text #15

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.2978. Min Act: -0.0000

Data Index: 961141 (C4 (Web Text))

Max Activating Token Index: 533

Text #16

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.2956. Min Act: -0.0000

Data Index: 1607992 (Python Code)

Max Activating Token Index: 113

Text #17

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.2214. Min Act: -0.0000

Data Index: 1297850 (C4 (Web Text))

Max Activating Token Index: 694

Text #18

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.2477. Min Act: -0.0000

Data Index: 1472837 (Python Code)

Max Activating Token Index: 157

Text #19

Max Range: 2.0358. Min Range: -2.0358

Max Act: 1.2738. Min Act: -0.0000

Data Index: 592990 (C4 (Web Text))

Max Activating Token Index: 178