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

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

Neuron 2585 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: 1.3857. Min Range: -1.3857

Max Act: 1.3857. Min Act: -0.0000

Data Index: 351486 (C4 (Web Text))

Max Activating Token Index: 887

Text #1

Max Range: 1.3857. Min Range: -1.3857

Max Act: 1.3681. Min Act: -0.0000

Data Index: 872309 (C4 (Web Text))

Max Activating Token Index: 283

Text #2

Max Range: 1.3857. Min Range: -1.3857

Max Act: 1.2732. Min Act: -0.0000

Data Index: 1031463 (C4 (Web Text))

Max Activating Token Index: 402

Text #3

Max Range: 1.3857. Min Range: -1.3857

Max Act: 1.2628. Min Act: -0.0000

Data Index: 1141372 (C4 (Web Text))

Max Activating Token Index: 100

Text #4

Max Range: 1.3857. Min Range: -1.3857

Max Act: 1.2795. Min Act: -0.0000

Data Index: 653574 (C4 (Web Text))

Max Activating Token Index: 1011

Text #5

Max Range: 1.3857. Min Range: -1.3857

Max Act: 1.1555. Min Act: -0.0000

Data Index: 740965 (C4 (Web Text))

Max Activating Token Index: 394

Text #6

Max Range: 1.3857. Min Range: -1.3857

Max Act: 1.0334. Min Act: -0.0000

Data Index: 79024 (C4 (Web Text))

Max Activating Token Index: 326

Text #7

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.9998. Min Act: -0.0000

Data Index: 1302750 (C4 (Web Text))

Max Activating Token Index: 652

Text #8

Max Range: 1.3857. Min Range: -1.3857

Max Act: 1.0256. Min Act: -0.0000

Data Index: 164301 (C4 (Web Text))

Max Activating Token Index: 323

Text #9

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.9533. Min Act: -0.0000

Data Index: 275224 (C4 (Web Text))

Max Activating Token Index: 28

Text #10

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.9426. Min Act: -0.0000

Data Index: 1066458 (C4 (Web Text))

Max Activating Token Index: 70

Text #11

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.9362. Min Act: -0.0000

Data Index: 641303 (C4 (Web Text))

Max Activating Token Index: 648

Text #12

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.8615. Min Act: -0.0000

Data Index: 9460 (C4 (Web Text))

Max Activating Token Index: 981

Text #13

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.8881. Min Act: -0.0000

Data Index: 326934 (C4 (Web Text))

Max Activating Token Index: 477

Text #14

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.9109. Min Act: -0.0000

Data Index: 1379191 (Python Code)

Max Activating Token Index: 917

Text #15

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.8758. Min Act: -0.0000

Data Index: 556855 (C4 (Web Text))

Max Activating Token Index: 785

Text #16

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.9035. Min Act: -0.0000

Data Index: 1142153 (C4 (Web Text))

Max Activating Token Index: 14

Text #17

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.8427. Min Act: -0.0000

Data Index: 906479 (C4 (Web Text))

Max Activating Token Index: 203

Text #18

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.8715. Min Act: -0.0000

Data Index: 673160 (C4 (Web Text))

Max Activating Token Index: 74

Text #19

Max Range: 1.3857. Min Range: -1.3857

Max Act: 0.9249. Min Act: -0.0000

Data Index: 819202 (C4 (Web Text))

Max Activating Token Index: 431