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

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

Neuron 4975 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.6387. Min Act: -0.0000

Data Index: 1384967 (Python Code)

Max Activating Token Index: 937

Text #1

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.5804. Min Act: -0.0000

Data Index: 1582504 (Python Code)

Max Activating Token Index: 958

Text #2

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.4424. Min Act: -0.0000

Data Index: 563728 (C4 (Web Text))

Max Activating Token Index: 759

Text #3

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.3548. Min Act: -0.0000

Data Index: 233333 (C4 (Web Text))

Max Activating Token Index: 759

Text #4

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.0970. Min Act: -0.0000

Data Index: 31807 (C4 (Web Text))

Max Activating Token Index: 245

Text #5

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.0596. Min Act: -0.0000

Data Index: 1524148 (Python Code)

Max Activating Token Index: 194

Text #6

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.0903. Min Act: -0.0000

Data Index: 139532 (C4 (Web Text))

Max Activating Token Index: 456

Text #7

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.0971. Min Act: -0.0000

Data Index: 360131 (C4 (Web Text))

Max Activating Token Index: 252

Text #8

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.0727. Min Act: -0.0000

Data Index: 1623258 (Python Code)

Max Activating Token Index: 357

Text #9

Max Range: 1.6387. Min Range: -1.6387

Max Act: 1.0726. Min Act: -0.0000

Data Index: 1555488 (Python Code)

Max Activating Token Index: 870

Text #10

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.9431. Min Act: -0.0000

Data Index: 896601 (C4 (Web Text))

Max Activating Token Index: 1004

Text #11

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.9983. Min Act: -0.0000

Data Index: 103075 (C4 (Web Text))

Max Activating Token Index: 701

Text #12

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.9287. Min Act: -0.0000

Data Index: 1460279 (Python Code)

Max Activating Token Index: 361

Text #13

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.8693. Min Act: -0.0000

Data Index: 1541979 (Python Code)

Max Activating Token Index: 684

Text #14

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.8847. Min Act: -0.0000

Data Index: 1552444 (Python Code)

Max Activating Token Index: 160

Text #15

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.8534. Min Act: -0.0000

Data Index: 731615 (C4 (Web Text))

Max Activating Token Index: 149

Text #16

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.8404. Min Act: -0.0000

Data Index: 857381 (C4 (Web Text))

Max Activating Token Index: 738

Text #17

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.8337. Min Act: -0.0000

Data Index: 265649 (C4 (Web Text))

Max Activating Token Index: 322

Text #18

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.8563. Min Act: -0.0000

Data Index: 1073671 (C4 (Web Text))

Max Activating Token Index: 405

Text #19

Max Range: 1.6387. Min Range: -1.6387

Max Act: 0.8013. Min Act: -0.0000

Data Index: 1298535 (C4 (Web Text))

Max Activating Token Index: 694