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

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

Neuron 753 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: 0.5772. Min Range: -0.5772

Max Act: 0.5772. Min Act: -0.0000

Data Index: 297525 (C4 (Web Text))

Max Activating Token Index: 995

Text #1

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4980. Min Act: -0.0000

Data Index: 1001573 (C4 (Web Text))

Max Activating Token Index: 155

Text #2

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.3538. Min Act: -0.0000

Data Index: 356494 (C4 (Web Text))

Max Activating Token Index: 622

Text #3

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.3291. Min Act: -0.0000

Data Index: 597328 (C4 (Web Text))

Max Activating Token Index: 808

Text #4

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.3356. Min Act: -0.0000

Data Index: 544602 (C4 (Web Text))

Max Activating Token Index: 539

Text #5

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2947. Min Act: -0.0000

Data Index: 354191 (C4 (Web Text))

Max Activating Token Index: 981

Text #6

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.3020. Min Act: -0.0000

Data Index: 19650 (C4 (Web Text))

Max Activating Token Index: 27

Text #7

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.3079. Min Act: -0.0000

Data Index: 630053 (C4 (Web Text))

Max Activating Token Index: 802

Text #8

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.3063. Min Act: -0.0000

Data Index: 352890 (C4 (Web Text))

Max Activating Token Index: 980

Text #9

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2629. Min Act: -0.0000

Data Index: 28202 (C4 (Web Text))

Max Activating Token Index: 464

Text #10

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2678. Min Act: -0.0000

Data Index: 594885 (C4 (Web Text))

Max Activating Token Index: 993

Text #11

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2772. Min Act: -0.0000

Data Index: 710266 (C4 (Web Text))

Max Activating Token Index: 87

Text #12

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2771. Min Act: -0.0000

Data Index: 44467 (C4 (Web Text))

Max Activating Token Index: 848

Text #13

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2859. Min Act: -0.0000

Data Index: 793841 (C4 (Web Text))

Max Activating Token Index: 620

Text #14

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2653. Min Act: -0.0000

Data Index: 1469322 (Python Code)

Max Activating Token Index: 878

Text #15

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2671. Min Act: -0.0000

Data Index: 327322 (C4 (Web Text))

Max Activating Token Index: 928

Text #16

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2544. Min Act: -0.0000

Data Index: 1306859 (C4 (Web Text))

Max Activating Token Index: 539

Text #17

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2511. Min Act: -0.0000

Data Index: 1017467 (C4 (Web Text))

Max Activating Token Index: 646

Text #18

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2240. Min Act: -0.0000

Data Index: 966666 (C4 (Web Text))

Max Activating Token Index: 295

Text #19

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.2525. Min Act: -0.0000

Data Index: 1251606 (C4 (Web Text))

Max Activating Token Index: 712