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

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

Neuron 5795 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.4600. Min Act: -0.0001

Data Index: 61328 (C4 (Web Text))

Max Activating Token Index: 848

Text #1

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.1359. Min Act: -0.0001

Data Index: 736851 (C4 (Web Text))

Max Activating Token Index: 439

Text #2

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.1708. Min Act: -0.0001

Data Index: 571423 (C4 (Web Text))

Max Activating Token Index: 386

Text #3

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.3132. Min Act: -0.0001

Data Index: 594903 (C4 (Web Text))

Max Activating Token Index: 766

Text #4

Max Range: 3.4600. Min Range: -3.4600

Max Act: 2.9996. Min Act: -0.0000

Data Index: 1497307 (Python Code)

Max Activating Token Index: 347

Text #5

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.3392. Min Act: -0.0001

Data Index: 676982 (C4 (Web Text))

Max Activating Token Index: 766

Text #6

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.2402. Min Act: -0.0000

Data Index: 1558381 (Python Code)

Max Activating Token Index: 118

Text #7

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.1670. Min Act: -0.0001

Data Index: 988914 (C4 (Web Text))

Max Activating Token Index: 108

Text #8

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.2446. Min Act: -0.0000

Data Index: 1594585 (Python Code)

Max Activating Token Index: 105

Text #9

Max Range: 3.4600. Min Range: -3.4600

Max Act: 2.9408. Min Act: -0.0001

Data Index: 289715 (C4 (Web Text))

Max Activating Token Index: 768

Text #10

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.0761. Min Act: -0.0001

Data Index: 723258 (C4 (Web Text))

Max Activating Token Index: 462

Text #11

Max Range: 3.4600. Min Range: -3.4600

Max Act: 2.9247. Min Act: -0.0001

Data Index: 1403115 (Python Code)

Max Activating Token Index: 851

Text #12

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.0747. Min Act: -0.0000

Data Index: 1464390 (Python Code)

Max Activating Token Index: 85

Text #13

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.0745. Min Act: -0.0001

Data Index: 1655021 (Python Code)

Max Activating Token Index: 19

Text #14

Max Range: 3.4600. Min Range: -3.4600

Max Act: 2.7583. Min Act: -0.0000

Data Index: 1604167 (Python Code)

Max Activating Token Index: 108

Text #15

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.2998. Min Act: -0.0001

Data Index: 1616361 (Python Code)

Max Activating Token Index: 817

Text #16

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.1387. Min Act: -0.0001

Data Index: 591100 (C4 (Web Text))

Max Activating Token Index: 976

Text #17

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.4332. Min Act: -0.0001

Data Index: 1306984 (C4 (Web Text))

Max Activating Token Index: 937

Text #18

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.0140. Min Act: -0.0001

Data Index: 1492158 (Python Code)

Max Activating Token Index: 437

Text #19

Max Range: 3.4600. Min Range: -3.4600

Max Act: 3.1591. Min Act: -0.0001

Data Index: 69098 (C4 (Web Text))

Max Activating Token Index: 899