Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4103 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.1414. Min Act: -0.0000

Data Index: 692401 (C4 (Web Text))

Max Activating Token Index: 617

Text #1

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.3590. Min Act: -0.0000

Data Index: 860517 (C4 (Web Text))

Max Activating Token Index: 935

Text #2

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.2099. Min Act: -0.0000

Data Index: 302486 (C4 (Web Text))

Max Activating Token Index: 207

Text #3

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.1026. Min Act: -0.0000

Data Index: 61844 (C4 (Web Text))

Max Activating Token Index: 446

Text #4

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.1378. Min Act: -0.0000

Data Index: 1262645 (C4 (Web Text))

Max Activating Token Index: 211

Text #5

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.0852. Min Act: -0.0001

Data Index: 742410 (C4 (Web Text))

Max Activating Token Index: 81

Text #6

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.0867. Min Act: -0.0000

Data Index: 1117734 (C4 (Web Text))

Max Activating Token Index: 763

Text #7

Max Range: 2.5498. Min Range: -2.5498

Max Act: 1.9761. Min Act: -0.0000

Data Index: 1424398 (Python Code)

Max Activating Token Index: 53

Text #8

Max Range: 2.5498. Min Range: -2.5498

Max Act: 1.8440. Min Act: -0.0000

Data Index: 208553 (C4 (Web Text))

Max Activating Token Index: 343

Text #9

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.1358. Min Act: -0.0001

Data Index: 35784 (C4 (Web Text))

Max Activating Token Index: 193

Text #10

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.0168. Min Act: -0.0000

Data Index: 1025791 (C4 (Web Text))

Max Activating Token Index: 35

Text #11

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.5498. Min Act: -0.0001

Data Index: 1491387 (Python Code)

Max Activating Token Index: 1003

Text #12

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.0899. Min Act: -0.0000

Data Index: 1225344 (C4 (Web Text))

Max Activating Token Index: 643

Text #13

Max Range: 2.5498. Min Range: -2.5498

Max Act: 2.0531. Min Act: -0.0000

Data Index: 1501878 (Python Code)

Max Activating Token Index: 255

Text #14

Max Range: 2.5498. Min Range: -2.5498

Max Act: 1.8083. Min Act: -0.0001

Data Index: 1428637 (Python Code)

Max Activating Token Index: 807

Text #15

Max Range: 2.5498. Min Range: -2.5498

Max Act: 1.8296. Min Act: -0.0000

Data Index: 704636 (C4 (Web Text))

Max Activating Token Index: 572

Text #16

Max Range: 2.5498. Min Range: -2.5498

Max Act: 1.9631. Min Act: -0.0000

Data Index: 147617 (C4 (Web Text))

Max Activating Token Index: 252

Text #17

Max Range: 2.5498. Min Range: -2.5498

Max Act: 1.9859. Min Act: -0.0000

Data Index: 1014446 (C4 (Web Text))

Max Activating Token Index: 90

Text #18

Max Range: 2.5498. Min Range: -2.5498

Max Act: 1.7471. Min Act: -0.0000

Data Index: 69100 (C4 (Web Text))

Max Activating Token Index: 862

Text #19

Max Range: 2.5498. Min Range: -2.5498

Max Act: 1.5990. Min Act: -0.0000

Data Index: 1405930 (Python Code)

Max Activating Token Index: 663