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

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

Neuron 1964 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: 2.7757. Min Range: -2.7757

Max Act: 2.7757. Min Act: -0.0000

Data Index: 1493622 (Python Code)

Max Activating Token Index: 928

Text #1

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.6743. Min Act: -0.0000

Data Index: 1540417 (Python Code)

Max Activating Token Index: 950

Text #2

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.4856. Min Act: -0.0000

Data Index: 1471027 (Python Code)

Max Activating Token Index: 917

Text #3

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.5242. Min Act: -0.0000

Data Index: 566769 (C4 (Web Text))

Max Activating Token Index: 565

Text #4

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.6538. Min Act: -0.0000

Data Index: 1455466 (Python Code)

Max Activating Token Index: 948

Text #5

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.3505. Min Act: -0.0000

Data Index: 169867 (C4 (Web Text))

Max Activating Token Index: 723

Text #6

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.4125. Min Act: -0.0000

Data Index: 1499063 (Python Code)

Max Activating Token Index: 275

Text #7

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.4105. Min Act: -0.0000

Data Index: 1497607 (Python Code)

Max Activating Token Index: 501

Text #8

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.3798. Min Act: -0.0000

Data Index: 863684 (C4 (Web Text))

Max Activating Token Index: 218

Text #9

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.3273. Min Act: -0.0000

Data Index: 1463464 (Python Code)

Max Activating Token Index: 387

Text #10

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.3278. Min Act: -0.0000

Data Index: 839787 (C4 (Web Text))

Max Activating Token Index: 383

Text #11

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.2694. Min Act: -0.0000

Data Index: 1534530 (Python Code)

Max Activating Token Index: 508

Text #12

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.2711. Min Act: -0.0000

Data Index: 1566149 (Python Code)

Max Activating Token Index: 494

Text #13

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.3332. Min Act: -0.0000

Data Index: 1564208 (Python Code)

Max Activating Token Index: 823

Text #14

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.2759. Min Act: -0.0000

Data Index: 1508592 (Python Code)

Max Activating Token Index: 110

Text #15

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.1765. Min Act: -0.0000

Data Index: 1339430 (C4 (Web Text))

Max Activating Token Index: 108

Text #16

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.2479. Min Act: -0.0000

Data Index: 1512509 (Python Code)

Max Activating Token Index: 702

Text #17

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.1457. Min Act: -0.0000

Data Index: 343839 (C4 (Web Text))

Max Activating Token Index: 621

Text #18

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.0588. Min Act: -0.0000

Data Index: 386300 (C4 (Web Text))

Max Activating Token Index: 60

Text #19

Max Range: 2.7757. Min Range: -2.7757

Max Act: 2.0487. Min Act: -0.0000

Data Index: 253969 (C4 (Web Text))

Max Activating Token Index: 22