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

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

Neuron 1996 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.0937. Min Range: -4.0937

Max Act: 4.0566. Min Act: -0.0000

Data Index: 1589132 (Python Code)

Max Activating Token Index: 25

Text #1

Max Range: 4.0937. Min Range: -4.0937

Max Act: 4.0937. Min Act: -0.0000

Data Index: 768437 (C4 (Web Text))

Max Activating Token Index: 457

Text #2

Max Range: 4.0937. Min Range: -4.0937

Max Act: 3.9999. Min Act: -0.0000

Data Index: 1075773 (C4 (Web Text))

Max Activating Token Index: 245

Text #3

Max Range: 4.0937. Min Range: -4.0937

Max Act: 3.5385. Min Act: -0.0000

Data Index: 1462419 (Python Code)

Max Activating Token Index: 250

Text #4

Max Range: 4.0937. Min Range: -4.0937

Max Act: 3.3375. Min Act: -0.0000

Data Index: 383435 (C4 (Web Text))

Max Activating Token Index: 663

Text #5

Max Range: 4.0937. Min Range: -4.0937

Max Act: 3.0048. Min Act: -0.0000

Data Index: 670803 (C4 (Web Text))

Max Activating Token Index: 998

Text #6

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.8543. Min Act: -0.0000

Data Index: 515363 (C4 (Web Text))

Max Activating Token Index: 965

Text #7

Max Range: 4.0937. Min Range: -4.0937

Max Act: 3.0904. Min Act: -0.0000

Data Index: 548138 (C4 (Web Text))

Max Activating Token Index: 73

Text #8

Max Range: 4.0937. Min Range: -4.0937

Max Act: 3.2828. Min Act: -0.0000

Data Index: 1334476 (C4 (Web Text))

Max Activating Token Index: 895

Text #9

Max Range: 4.0937. Min Range: -4.0937

Max Act: 3.0070. Min Act: -0.0000

Data Index: 664323 (C4 (Web Text))

Max Activating Token Index: 942

Text #10

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.8787. Min Act: -0.0000

Data Index: 12341 (C4 (Web Text))

Max Activating Token Index: 356

Text #11

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.7789. Min Act: -0.0000

Data Index: 907350 (C4 (Web Text))

Max Activating Token Index: 19

Text #12

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.7684. Min Act: -0.0000

Data Index: 1513890 (Python Code)

Max Activating Token Index: 369

Text #13

Max Range: 4.0937. Min Range: -4.0937

Max Act: 3.0442. Min Act: -0.0000

Data Index: 56582 (C4 (Web Text))

Max Activating Token Index: 23

Text #14

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.7386. Min Act: -0.0000

Data Index: 59278 (C4 (Web Text))

Max Activating Token Index: 76

Text #15

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.5972. Min Act: -0.0000

Data Index: 1556053 (Python Code)

Max Activating Token Index: 763

Text #16

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.6406. Min Act: -0.0000

Data Index: 792929 (C4 (Web Text))

Max Activating Token Index: 548

Text #17

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.4515. Min Act: -0.0000

Data Index: 357993 (C4 (Web Text))

Max Activating Token Index: 321

Text #18

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.7320. Min Act: -0.0000

Data Index: 931783 (C4 (Web Text))

Max Activating Token Index: 212

Text #19

Max Range: 4.0937. Min Range: -4.0937

Max Act: 2.5789. Min Act: -0.0000

Data Index: 442773 (C4 (Web Text))

Max Activating Token Index: 540