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

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

Neuron 530 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: 0.6825. Min Range: -0.6825

Max Act: 0.6825. Min Act: -0.0000

Data Index: 613584 (C4 (Web Text))

Max Activating Token Index: 130

Text #1

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.6133. Min Act: -0.0000

Data Index: 964755 (C4 (Web Text))

Max Activating Token Index: 768

Text #2

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.5937. Min Act: -0.0000

Data Index: 1336589 (C4 (Web Text))

Max Activating Token Index: 409

Text #3

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.5204. Min Act: -0.0001

Data Index: 29976 (C4 (Web Text))

Max Activating Token Index: 15

Text #4

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.4786. Min Act: -0.0001

Data Index: 171146 (C4 (Web Text))

Max Activating Token Index: 277

Text #5

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.4059. Min Act: -0.0000

Data Index: 619070 (C4 (Web Text))

Max Activating Token Index: 468

Text #6

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3711. Min Act: -0.0001

Data Index: 481730 (C4 (Web Text))

Max Activating Token Index: 258

Text #7

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3429. Min Act: -0.0000

Data Index: 696349 (C4 (Web Text))

Max Activating Token Index: 520

Text #8

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3248. Min Act: -0.0001

Data Index: 1576714 (Python Code)

Max Activating Token Index: 20

Text #9

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3345. Min Act: -0.0001

Data Index: 383911 (C4 (Web Text))

Max Activating Token Index: 93

Text #10

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3086. Min Act: -0.0001

Data Index: 77148 (C4 (Web Text))

Max Activating Token Index: 799

Text #11

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3269. Min Act: -0.0001

Data Index: 972131 (C4 (Web Text))

Max Activating Token Index: 22

Text #12

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3017. Min Act: -0.0001

Data Index: 920967 (C4 (Web Text))

Max Activating Token Index: 938

Text #13

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3204. Min Act: -0.0001

Data Index: 763302 (C4 (Web Text))

Max Activating Token Index: 180

Text #14

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3179. Min Act: -0.0000

Data Index: 37371 (C4 (Web Text))

Max Activating Token Index: 8

Text #15

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3038. Min Act: -0.0001

Data Index: 97847 (C4 (Web Text))

Max Activating Token Index: 405

Text #16

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3137. Min Act: -0.0001

Data Index: 786082 (C4 (Web Text))

Max Activating Token Index: 399

Text #17

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.2907. Min Act: -0.0001

Data Index: 30840 (C4 (Web Text))

Max Activating Token Index: 720

Text #18

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.2956. Min Act: -0.0001

Data Index: 757972 (C4 (Web Text))

Max Activating Token Index: 210

Text #19

Max Range: 0.6825. Min Range: -0.6825

Max Act: 0.3043. Min Act: -0.0000

Data Index: 1190651 (C4 (Web Text))

Max Activating Token Index: 10