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

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

Neuron 4586 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.3715. Min Act: -0.0000

Data Index: 464022 (C4 (Web Text))

Max Activating Token Index: 840

Text #1

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.3480. Min Act: -0.0000

Data Index: 277008 (C4 (Web Text))

Max Activating Token Index: 946

Text #2

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.3222. Min Act: -0.0000

Data Index: 851196 (C4 (Web Text))

Max Activating Token Index: 75

Text #3

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2932. Min Act: -0.0000

Data Index: 585412 (C4 (Web Text))

Max Activating Token Index: 522

Text #4

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2766. Min Act: -0.0000

Data Index: 1467770 (Python Code)

Max Activating Token Index: 387

Text #5

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.3154. Min Act: -0.0000

Data Index: 1003172 (C4 (Web Text))

Max Activating Token Index: 917

Text #6

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2741. Min Act: -0.0000

Data Index: 530774 (C4 (Web Text))

Max Activating Token Index: 800

Text #7

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2753. Min Act: -0.0000

Data Index: 639696 (C4 (Web Text))

Max Activating Token Index: 982

Text #8

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2494. Min Act: -0.0000

Data Index: 1133777 (C4 (Web Text))

Max Activating Token Index: 80

Text #9

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2601. Min Act: -0.0000

Data Index: 436232 (C4 (Web Text))

Max Activating Token Index: 638

Text #10

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2472. Min Act: -0.0000

Data Index: 1251405 (C4 (Web Text))

Max Activating Token Index: 265

Text #11

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2121. Min Act: -0.0000

Data Index: 1411487 (Python Code)

Max Activating Token Index: 430

Text #12

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2428. Min Act: -0.0000

Data Index: 982894 (C4 (Web Text))

Max Activating Token Index: 344

Text #13

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2295. Min Act: -0.0000

Data Index: 1193376 (C4 (Web Text))

Max Activating Token Index: 994

Text #14

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2128. Min Act: -0.0000

Data Index: 233869 (C4 (Web Text))

Max Activating Token Index: 324

Text #15

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2059. Min Act: -0.0000

Data Index: 438515 (C4 (Web Text))

Max Activating Token Index: 194

Text #16

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2203. Min Act: -0.0000

Data Index: 2253 (C4 (Web Text))

Max Activating Token Index: 91

Text #17

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2125. Min Act: -0.0000

Data Index: 1339742 (C4 (Web Text))

Max Activating Token Index: 165

Text #18

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2056. Min Act: -0.0000

Data Index: 1226159 (C4 (Web Text))

Max Activating Token Index: 126

Text #19

Max Range: 0.3715. Min Range: -0.3715

Max Act: 0.2045. Min Act: -0.0000

Data Index: 1330776 (C4 (Web Text))

Max Activating Token Index: 813