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

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

Neuron 2448 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.9704. Min Act: -0.0000

Data Index: 1588960 (Python Code)

Max Activating Token Index: 981

Text #1

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.8813. Min Act: -0.0000

Data Index: 1303013 (C4 (Web Text))

Max Activating Token Index: 235

Text #2

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.8566. Min Act: -0.0000

Data Index: 1490637 (Python Code)

Max Activating Token Index: 795

Text #3

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.8405. Min Act: -0.0000

Data Index: 1493711 (Python Code)

Max Activating Token Index: 200

Text #4

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.8700. Min Act: -0.0000

Data Index: 1472339 (Python Code)

Max Activating Token Index: 888

Text #5

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.7441. Min Act: -0.0000

Data Index: 1236777 (C4 (Web Text))

Max Activating Token Index: 361

Text #6

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.7422. Min Act: -0.0000

Data Index: 573189 (C4 (Web Text))

Max Activating Token Index: 740

Text #7

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.7105. Min Act: -0.0000

Data Index: 1432259 (Python Code)

Max Activating Token Index: 436

Text #8

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.6848. Min Act: -0.0000

Data Index: 849410 (C4 (Web Text))

Max Activating Token Index: 792

Text #9

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.6289. Min Act: -0.0000

Data Index: 983984 (C4 (Web Text))

Max Activating Token Index: 64

Text #10

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.5921. Min Act: -0.0000

Data Index: 1583204 (Python Code)

Max Activating Token Index: 429

Text #11

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.5866. Min Act: -0.0000

Data Index: 1634416 (Python Code)

Max Activating Token Index: 71

Text #12

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.5389. Min Act: -0.0000

Data Index: 811723 (C4 (Web Text))

Max Activating Token Index: 901

Text #13

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.5282. Min Act: -0.0000

Data Index: 1645598 (Python Code)

Max Activating Token Index: 620

Text #14

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.5344. Min Act: -0.0000

Data Index: 1019513 (C4 (Web Text))

Max Activating Token Index: 344

Text #15

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.5276. Min Act: -0.0000

Data Index: 742782 (C4 (Web Text))

Max Activating Token Index: 929

Text #16

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.5297. Min Act: -0.0000

Data Index: 1379854 (Python Code)

Max Activating Token Index: 425

Text #17

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.5261. Min Act: -0.0000

Data Index: 1611187 (Python Code)

Max Activating Token Index: 357

Text #18

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.4897. Min Act: -0.0000

Data Index: 1597311 (Python Code)

Max Activating Token Index: 943

Text #19

Max Range: 0.9704. Min Range: -0.9704

Max Act: 0.4842. Min Act: -0.0000

Data Index: 1471140 (Python Code)

Max Activating Token Index: 495