Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1092 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0387. Min Act: -0.0001

Data Index: 933735 (C4 (Web Text))

Max Activating Token Index: 401

Text #1

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0327. Min Act: -0.0001

Data Index: 367490 (C4 (Web Text))

Max Activating Token Index: 387

Text #2

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0287. Min Act: -0.0001

Data Index: 294134 (C4 (Web Text))

Max Activating Token Index: 210

Text #3

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0261. Min Act: -0.0001

Data Index: 654718 (C4 (Web Text))

Max Activating Token Index: 862

Text #4

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0243. Min Act: -0.0001

Data Index: 565123 (C4 (Web Text))

Max Activating Token Index: 105

Text #5

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0243. Min Act: -0.0001

Data Index: 817623 (C4 (Web Text))

Max Activating Token Index: 509

Text #6

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0219. Min Act: -0.0001

Data Index: 721417 (C4 (Web Text))

Max Activating Token Index: 219

Text #7

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0209. Min Act: -0.0001

Data Index: 646209 (C4 (Web Text))

Max Activating Token Index: 659

Text #8

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0214. Min Act: -0.0001

Data Index: 1586175 (Python Code)

Max Activating Token Index: 268

Text #9

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0204. Min Act: -0.0001

Data Index: 937656 (C4 (Web Text))

Max Activating Token Index: 999

Text #10

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0208. Min Act: -0.0001

Data Index: 263833 (C4 (Web Text))

Max Activating Token Index: 190

Text #11

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0197. Min Act: -0.0001

Data Index: 1109705 (C4 (Web Text))

Max Activating Token Index: 598

Text #12

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0202. Min Act: -0.0001

Data Index: 536807 (C4 (Web Text))

Max Activating Token Index: 972

Text #13

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0183. Min Act: -0.0000

Data Index: 966082 (C4 (Web Text))

Max Activating Token Index: 392

Text #14

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0182. Min Act: -0.0001

Data Index: 1370045 (Python Code)

Max Activating Token Index: 716

Text #15

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0168. Min Act: -0.0001

Data Index: 484795 (C4 (Web Text))

Max Activating Token Index: 256

Text #16

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0181. Min Act: -0.0001

Data Index: 1526976 (Python Code)

Max Activating Token Index: 660

Text #17

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0200. Min Act: -0.0001

Data Index: 1393136 (Python Code)

Max Activating Token Index: 754

Text #18

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0186. Min Act: -0.0001

Data Index: 873901 (C4 (Web Text))

Max Activating Token Index: 1022

Text #19

Max Range: 0.0387. Min Range: -0.0387

Max Act: 0.0172. Min Act: -0.0001

Data Index: 1548596 (Python Code)

Max Activating Token Index: 240