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

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

Neuron 135 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3154. Min Range: -1.3154

Max Act: 1.3154. Min Act: -0.0000

Data Index: 1228743 (C4 (Web Text))

Max Activating Token Index: 133

Text #1

Max Range: 1.3154. Min Range: -1.3154

Max Act: 1.1751. Min Act: -0.0000

Data Index: 428141 (C4 (Web Text))

Max Activating Token Index: 472

Text #2

Max Range: 1.3154. Min Range: -1.3154

Max Act: 1.0575. Min Act: -0.0000

Data Index: 1143594 (C4 (Web Text))

Max Activating Token Index: 26

Text #3

Max Range: 1.3154. Min Range: -1.3154

Max Act: 1.0357. Min Act: -0.0000

Data Index: 1058630 (C4 (Web Text))

Max Activating Token Index: 174

Text #4

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.9938. Min Act: -0.0000

Data Index: 568430 (C4 (Web Text))

Max Activating Token Index: 79

Text #5

Max Range: 1.3154. Min Range: -1.3154

Max Act: 1.0020. Min Act: -0.0000

Data Index: 975495 (C4 (Web Text))

Max Activating Token Index: 12

Text #6

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.9777. Min Act: -0.0000

Data Index: 683179 (C4 (Web Text))

Max Activating Token Index: 602

Text #7

Max Range: 1.3154. Min Range: -1.3154

Max Act: 1.0290. Min Act: -0.0000

Data Index: 82485 (C4 (Web Text))

Max Activating Token Index: 511

Text #8

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.9052. Min Act: -0.0000

Data Index: 352916 (C4 (Web Text))

Max Activating Token Index: 409

Text #9

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.9023. Min Act: -0.0000

Data Index: 506163 (C4 (Web Text))

Max Activating Token Index: 809

Text #10

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.8992. Min Act: -0.0000

Data Index: 1301102 (C4 (Web Text))

Max Activating Token Index: 9

Text #11

Max Range: 1.3154. Min Range: -1.3154

Max Act: 1.0009. Min Act: -0.0000

Data Index: 1588396 (Python Code)

Max Activating Token Index: 815

Text #12

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.8174. Min Act: -0.0000

Data Index: 643523 (C4 (Web Text))

Max Activating Token Index: 612

Text #13

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.9219. Min Act: -0.0000

Data Index: 1463429 (Python Code)

Max Activating Token Index: 34

Text #14

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.8212. Min Act: -0.0000

Data Index: 6917 (C4 (Web Text))

Max Activating Token Index: 805

Text #15

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.8374. Min Act: -0.0000

Data Index: 268772 (C4 (Web Text))

Max Activating Token Index: 441

Text #16

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.8657. Min Act: -0.0000

Data Index: 1053974 (C4 (Web Text))

Max Activating Token Index: 1017

Text #17

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.8370. Min Act: -0.0000

Data Index: 554482 (C4 (Web Text))

Max Activating Token Index: 463

Text #18

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.8788. Min Act: -0.0000

Data Index: 956766 (C4 (Web Text))

Max Activating Token Index: 109

Text #19

Max Range: 1.3154. Min Range: -1.3154

Max Act: 0.7543. Min Act: -0.0000

Data Index: 1070145 (C4 (Web Text))

Max Activating Token Index: 937