Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 481 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0483. Min Act: -0.0003

Data Index: 95006 (C4 (Web Text))

Max Activating Token Index: 377

Text #1

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0472. Min Act: -0.0003

Data Index: 1461327 (Python Code)

Max Activating Token Index: 508

Text #2

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0473. Min Act: -0.0003

Data Index: 1497744 (Python Code)

Max Activating Token Index: 166

Text #3

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0447. Min Act: -0.0003

Data Index: 1058818 (C4 (Web Text))

Max Activating Token Index: 421

Text #4

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0471. Min Act: -0.0003

Data Index: 1621712 (Python Code)

Max Activating Token Index: 736

Text #5

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0431. Min Act: -0.0002

Data Index: 490434 (C4 (Web Text))

Max Activating Token Index: 734

Text #6

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0444. Min Act: -0.0004

Data Index: 1255384 (C4 (Web Text))

Max Activating Token Index: 485

Text #7

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0439. Min Act: -0.0003

Data Index: 1517481 (Python Code)

Max Activating Token Index: 869

Text #8

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0423. Min Act: -0.0003

Data Index: 186491 (C4 (Web Text))

Max Activating Token Index: 474

Text #9

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0389. Min Act: -0.0003

Data Index: 942687 (C4 (Web Text))

Max Activating Token Index: 314

Text #10

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0427. Min Act: -0.0003

Data Index: 521397 (C4 (Web Text))

Max Activating Token Index: 161

Text #11

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0416. Min Act: -0.0003

Data Index: 467557 (C4 (Web Text))

Max Activating Token Index: 537

Text #12

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0405. Min Act: -0.0003

Data Index: 1386919 (Python Code)

Max Activating Token Index: 737

Text #13

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0410. Min Act: -0.0003

Data Index: 1493534 (Python Code)

Max Activating Token Index: 894

Text #14

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0416. Min Act: -0.0003

Data Index: 1581642 (Python Code)

Max Activating Token Index: 604

Text #15

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0430. Min Act: -0.0003

Data Index: 538085 (C4 (Web Text))

Max Activating Token Index: 215

Text #16

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0409. Min Act: -0.0003

Data Index: 1394950 (Python Code)

Max Activating Token Index: 585

Text #17

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0402. Min Act: -0.0004

Data Index: 888012 (C4 (Web Text))

Max Activating Token Index: 370

Text #18

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0416. Min Act: -0.0003

Data Index: 783376 (C4 (Web Text))

Max Activating Token Index: 212

Text #19

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0407. Min Act: -0.0003

Data Index: 346496 (C4 (Web Text))

Max Activating Token Index: 937