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

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

Neuron 2187 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.6504. Min Act: -0.0000

Data Index: 1544104 (Python Code)

Max Activating Token Index: 60

Text #1

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.5617. Min Act: -0.0000

Data Index: 1465043 (Python Code)

Max Activating Token Index: 230

Text #2

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.4705. Min Act: -0.0000

Data Index: 482424 (C4 (Web Text))

Max Activating Token Index: 31

Text #3

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.3481. Min Act: -0.0000

Data Index: 1421588 (Python Code)

Max Activating Token Index: 22

Text #4

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.3337. Min Act: -0.0000

Data Index: 1471888 (Python Code)

Max Activating Token Index: 222

Text #5

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.2478. Min Act: -0.0000

Data Index: 589270 (C4 (Web Text))

Max Activating Token Index: 838

Text #6

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.2313. Min Act: -0.0000

Data Index: 1096310 (C4 (Web Text))

Max Activating Token Index: 834

Text #7

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.1719. Min Act: -0.0000

Data Index: 1499688 (Python Code)

Max Activating Token Index: 913

Text #8

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.1573. Min Act: -0.0000

Data Index: 78643 (C4 (Web Text))

Max Activating Token Index: 509

Text #9

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.2364. Min Act: -0.0000

Data Index: 584180 (C4 (Web Text))

Max Activating Token Index: 1020

Text #10

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.0024. Min Act: -0.0000

Data Index: 1622521 (Python Code)

Max Activating Token Index: 560

Text #11

Max Range: 2.6504. Min Range: -2.6504

Max Act: 1.9736. Min Act: -0.0000

Data Index: 1450215 (Python Code)

Max Activating Token Index: 429

Text #12

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.0760. Min Act: -0.0000

Data Index: 1399038 (Python Code)

Max Activating Token Index: 543

Text #13

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.0384. Min Act: -0.0000

Data Index: 636047 (C4 (Web Text))

Max Activating Token Index: 59

Text #14

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.0093. Min Act: -0.0000

Data Index: 1387690 (Python Code)

Max Activating Token Index: 347

Text #15

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.0974. Min Act: -0.0000

Data Index: 1252080 (C4 (Web Text))

Max Activating Token Index: 359

Text #16

Max Range: 2.6504. Min Range: -2.6504

Max Act: 1.9613. Min Act: -0.0000

Data Index: 1336195 (C4 (Web Text))

Max Activating Token Index: 901

Text #17

Max Range: 2.6504. Min Range: -2.6504

Max Act: 2.0326. Min Act: -0.0000

Data Index: 1487434 (Python Code)

Max Activating Token Index: 463

Text #18

Max Range: 2.6504. Min Range: -2.6504

Max Act: 1.9109. Min Act: -0.0000

Data Index: 487797 (C4 (Web Text))

Max Activating Token Index: 40

Text #19

Max Range: 2.6504. Min Range: -2.6504

Max Act: 1.8601. Min Act: -0.0000

Data Index: 646404 (C4 (Web Text))

Max Activating Token Index: 630