Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1886 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.3084. Min Range: -1.3084

Max Act: 1.3084. Min Act: -0.0001

Data Index: 563361 (C4 (Web Text))

Max Activating Token Index: 615

Text #1

Max Range: 1.3084. Min Range: -1.3084

Max Act: 1.2750. Min Act: -0.0001

Data Index: 1395597 (Python Code)

Max Activating Token Index: 530

Text #2

Max Range: 1.3084. Min Range: -1.3084

Max Act: 1.0888. Min Act: -0.0001

Data Index: 678280 (C4 (Web Text))

Max Activating Token Index: 235

Text #3

Max Range: 1.3084. Min Range: -1.3084

Max Act: 1.1967. Min Act: -0.0001

Data Index: 1401429 (Python Code)

Max Activating Token Index: 906

Text #4

Max Range: 1.3084. Min Range: -1.3084

Max Act: 1.2526. Min Act: -0.0001

Data Index: 860718 (C4 (Web Text))

Max Activating Token Index: 501

Text #5

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.9768. Min Act: -0.0001

Data Index: 21524 (C4 (Web Text))

Max Activating Token Index: 714

Text #6

Max Range: 1.3084. Min Range: -1.3084

Max Act: 1.0666. Min Act: -0.0001

Data Index: 42417 (C4 (Web Text))

Max Activating Token Index: 31

Text #7

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.9446. Min Act: -0.0001

Data Index: 1008717 (C4 (Web Text))

Max Activating Token Index: 626

Text #8

Max Range: 1.3084. Min Range: -1.3084

Max Act: 1.0429. Min Act: -0.0001

Data Index: 1129513 (C4 (Web Text))

Max Activating Token Index: 269

Text #9

Max Range: 1.3084. Min Range: -1.3084

Max Act: 1.0456. Min Act: -0.0001

Data Index: 828698 (C4 (Web Text))

Max Activating Token Index: 901

Text #10

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.9752. Min Act: -0.0001

Data Index: 921799 (C4 (Web Text))

Max Activating Token Index: 437

Text #11

Max Range: 1.3084. Min Range: -1.3084

Max Act: 1.0154. Min Act: -0.0001

Data Index: 1573935 (Python Code)

Max Activating Token Index: 345

Text #12

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.9566. Min Act: -0.0001

Data Index: 1557820 (Python Code)

Max Activating Token Index: 92

Text #13

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.9857. Min Act: -0.0001

Data Index: 1556718 (Python Code)

Max Activating Token Index: 9

Text #14

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.8945. Min Act: -0.0001

Data Index: 536987 (C4 (Web Text))

Max Activating Token Index: 300

Text #15

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.9461. Min Act: -0.0001

Data Index: 1512264 (Python Code)

Max Activating Token Index: 222

Text #16

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.9437. Min Act: -0.0001

Data Index: 965421 (C4 (Web Text))

Max Activating Token Index: 162

Text #17

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.8239. Min Act: -0.0001

Data Index: 1522341 (Python Code)

Max Activating Token Index: 733

Text #18

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.8986. Min Act: -0.0001

Data Index: 1410850 (Python Code)

Max Activating Token Index: 954

Text #19

Max Range: 1.3084. Min Range: -1.3084

Max Act: 0.8804. Min Act: -0.0001

Data Index: 329724 (C4 (Web Text))

Max Activating Token Index: 323