Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 963 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.4710. Min Act: -0.0001

Data Index: 1164395 (C4 (Web Text))

Max Activating Token Index: 918

Text #1

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.4356. Min Act: -0.0001

Data Index: 607486 (C4 (Web Text))

Max Activating Token Index: 449

Text #2

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.4068. Min Act: -0.0001

Data Index: 395358 (C4 (Web Text))

Max Activating Token Index: 278

Text #3

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3846. Min Act: -0.0001

Data Index: 239492 (C4 (Web Text))

Max Activating Token Index: 69

Text #4

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3839. Min Act: -0.0001

Data Index: 416896 (C4 (Web Text))

Max Activating Token Index: 27

Text #5

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3758. Min Act: -0.0001

Data Index: 1384034 (Python Code)

Max Activating Token Index: 692

Text #6

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3657. Min Act: -0.0001

Data Index: 1196486 (C4 (Web Text))

Max Activating Token Index: 670

Text #7

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3608. Min Act: -0.0000

Data Index: 1381979 (Python Code)

Max Activating Token Index: 709

Text #8

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3659. Min Act: -0.0000

Data Index: 718001 (C4 (Web Text))

Max Activating Token Index: 714

Text #9

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3660. Min Act: -0.0001

Data Index: 805600 (C4 (Web Text))

Max Activating Token Index: 29

Text #10

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3843. Min Act: -0.0001

Data Index: 210968 (C4 (Web Text))

Max Activating Token Index: 827

Text #11

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3582. Min Act: -0.0001

Data Index: 1277216 (C4 (Web Text))

Max Activating Token Index: 389

Text #12

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3759. Min Act: -0.0001

Data Index: 201181 (C4 (Web Text))

Max Activating Token Index: 285

Text #13

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3153. Min Act: -0.0000

Data Index: 1401145 (Python Code)

Max Activating Token Index: 335

Text #14

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3319. Min Act: -0.0001

Data Index: 1555130 (Python Code)

Max Activating Token Index: 611

Text #15

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3445. Min Act: -0.0001

Data Index: 1040871 (C4 (Web Text))

Max Activating Token Index: 34

Text #16

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3413. Min Act: -0.0001

Data Index: 810632 (C4 (Web Text))

Max Activating Token Index: 801

Text #17

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3407. Min Act: -0.0001

Data Index: 732455 (C4 (Web Text))

Max Activating Token Index: 802

Text #18

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3350. Min Act: -0.0001

Data Index: 1548374 (Python Code)

Max Activating Token Index: 654

Text #19

Max Range: 0.4710. Min Range: -0.4710

Max Act: 0.3460. Min Act: -0.0001

Data Index: 1115200 (C4 (Web Text))

Max Activating Token Index: 64