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

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

Neuron 514 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.4689. Min Act: -0.0000

Data Index: 1020217 (C4 (Web Text))

Max Activating Token Index: 1006

Text #1

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.0995. Min Act: -0.0000

Data Index: 790837 (C4 (Web Text))

Max Activating Token Index: 955

Text #2

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.1000. Min Act: -0.0000

Data Index: 656545 (C4 (Web Text))

Max Activating Token Index: 840

Text #3

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.0661. Min Act: -0.0000

Data Index: 23153 (C4 (Web Text))

Max Activating Token Index: 814

Text #4

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.0429. Min Act: -0.0000

Data Index: 794930 (C4 (Web Text))

Max Activating Token Index: 472

Text #5

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.0040. Min Act: -0.0000

Data Index: 667743 (C4 (Web Text))

Max Activating Token Index: 944

Text #6

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.0051. Min Act: -0.0000

Data Index: 299989 (C4 (Web Text))

Max Activating Token Index: 992

Text #7

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.0514. Min Act: 0.0000

Data Index: 775300 (C4 (Web Text))

Max Activating Token Index: 500

Text #8

Max Range: 1.4689. Min Range: -1.4689

Max Act: 0.9907. Min Act: -0.0000

Data Index: 831333 (C4 (Web Text))

Max Activating Token Index: 847

Text #9

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.0490. Min Act: 0.0000

Data Index: 1269368 (C4 (Web Text))

Max Activating Token Index: 194

Text #10

Max Range: 1.4689. Min Range: -1.4689

Max Act: 0.9759. Min Act: -0.0000

Data Index: 333348 (C4 (Web Text))

Max Activating Token Index: 519

Text #11

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.0355. Min Act: -0.0000

Data Index: 1085057 (C4 (Web Text))

Max Activating Token Index: 903

Text #12

Max Range: 1.4689. Min Range: -1.4689

Max Act: 0.9925. Min Act: -0.0000

Data Index: 1106417 (C4 (Web Text))

Max Activating Token Index: 869

Text #13

Max Range: 1.4689. Min Range: -1.4689

Max Act: 0.9954. Min Act: -0.0000

Data Index: 550627 (C4 (Web Text))

Max Activating Token Index: 150

Text #14

Max Range: 1.4689. Min Range: -1.4689

Max Act: 0.9852. Min Act: -0.0000

Data Index: 281180 (C4 (Web Text))

Max Activating Token Index: 654

Text #15

Max Range: 1.4689. Min Range: -1.4689

Max Act: 0.9791. Min Act: -0.0000

Data Index: 1458122 (Python Code)

Max Activating Token Index: 163

Text #16

Max Range: 1.4689. Min Range: -1.4689

Max Act: 1.0170. Min Act: -0.0000

Data Index: 180813 (C4 (Web Text))

Max Activating Token Index: 632

Text #17

Max Range: 1.4689. Min Range: -1.4689

Max Act: 0.9446. Min Act: -0.0000

Data Index: 911577 (C4 (Web Text))

Max Activating Token Index: 100

Text #18

Max Range: 1.4689. Min Range: -1.4689

Max Act: 0.9739. Min Act: -0.0000

Data Index: 1424945 (Python Code)

Max Activating Token Index: 461

Text #19

Max Range: 1.4689. Min Range: -1.4689

Max Act: 0.9964. Min Act: -0.0000

Data Index: 1290153 (C4 (Web Text))

Max Activating Token Index: 435