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

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

Neuron 4756 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.7601. Min Range: -4.7601

Max Act: 4.7601. Min Act: -0.0001

Data Index: 716378 (C4 (Web Text))

Max Activating Token Index: 660

Text #1

Max Range: 4.7601. Min Range: -4.7601

Max Act: 3.1430. Min Act: -0.0001

Data Index: 344592 (C4 (Web Text))

Max Activating Token Index: 51

Text #2

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.9923. Min Act: -0.0001

Data Index: 950000 (C4 (Web Text))

Max Activating Token Index: 972

Text #3

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.8405. Min Act: -0.0000

Data Index: 292620 (C4 (Web Text))

Max Activating Token Index: 12

Text #4

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.8127. Min Act: -0.0001

Data Index: 12992 (C4 (Web Text))

Max Activating Token Index: 980

Text #5

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.7733. Min Act: -0.0000

Data Index: 708020 (C4 (Web Text))

Max Activating Token Index: 278

Text #6

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.8903. Min Act: -0.0000

Data Index: 957715 (C4 (Web Text))

Max Activating Token Index: 62

Text #7

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.7399. Min Act: -0.0001

Data Index: 572734 (C4 (Web Text))

Max Activating Token Index: 101

Text #8

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.8402. Min Act: -0.0001

Data Index: 1047319 (C4 (Web Text))

Max Activating Token Index: 544

Text #9

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.7049. Min Act: -0.0001

Data Index: 132586 (C4 (Web Text))

Max Activating Token Index: 958

Text #10

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.7888. Min Act: -0.0001

Data Index: 1094879 (C4 (Web Text))

Max Activating Token Index: 29

Text #11

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.7533. Min Act: -0.0000

Data Index: 748258 (C4 (Web Text))

Max Activating Token Index: 49

Text #12

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.7322. Min Act: -0.0000

Data Index: 1148394 (C4 (Web Text))

Max Activating Token Index: 49

Text #13

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.7157. Min Act: -0.0001

Data Index: 1129309 (C4 (Web Text))

Max Activating Token Index: 156

Text #14

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.6903. Min Act: -0.0001

Data Index: 185698 (C4 (Web Text))

Max Activating Token Index: 90

Text #15

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.6707. Min Act: -0.0000

Data Index: 633822 (C4 (Web Text))

Max Activating Token Index: 60

Text #16

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.7977. Min Act: -0.0001

Data Index: 434873 (C4 (Web Text))

Max Activating Token Index: 104

Text #17

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.6457. Min Act: -0.0000

Data Index: 397382 (C4 (Web Text))

Max Activating Token Index: 46

Text #18

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.6431. Min Act: -0.0000

Data Index: 223282 (C4 (Web Text))

Max Activating Token Index: 49

Text #19

Max Range: 4.7601. Min Range: -4.7601

Max Act: 2.5177. Min Act: -0.0001

Data Index: 1173990 (C4 (Web Text))

Max Activating Token Index: 294