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

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

Neuron 683 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.4135. Min Range: -2.4135

Max Act: 2.4135. Min Act: -0.0000

Data Index: 345621 (C4 (Web Text))

Max Activating Token Index: 964

Text #1

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.9121. Min Act: -0.0000

Data Index: 403307 (C4 (Web Text))

Max Activating Token Index: 713

Text #2

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.4856. Min Act: -0.0000

Data Index: 814048 (C4 (Web Text))

Max Activating Token Index: 330

Text #3

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.5775. Min Act: -0.0000

Data Index: 1494346 (Python Code)

Max Activating Token Index: 241

Text #4

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.3793. Min Act: -0.0000

Data Index: 1229751 (C4 (Web Text))

Max Activating Token Index: 330

Text #5

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2891. Min Act: -0.0000

Data Index: 382674 (C4 (Web Text))

Max Activating Token Index: 64

Text #6

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2827. Min Act: -0.0000

Data Index: 448440 (C4 (Web Text))

Max Activating Token Index: 27

Text #7

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2663. Min Act: -0.0000

Data Index: 1433739 (Python Code)

Max Activating Token Index: 587

Text #8

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2045. Min Act: -0.0000

Data Index: 315015 (C4 (Web Text))

Max Activating Token Index: 300

Text #9

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.3081. Min Act: -0.0000

Data Index: 557656 (C4 (Web Text))

Max Activating Token Index: 90

Text #10

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2418. Min Act: -0.0000

Data Index: 1467004 (Python Code)

Max Activating Token Index: 251

Text #11

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2300. Min Act: -0.0000

Data Index: 1216413 (C4 (Web Text))

Max Activating Token Index: 207

Text #12

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2160. Min Act: -0.0000

Data Index: 712252 (C4 (Web Text))

Max Activating Token Index: 88

Text #13

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2286. Min Act: -0.0000

Data Index: 202349 (C4 (Web Text))

Max Activating Token Index: 906

Text #14

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2996. Min Act: -0.0000

Data Index: 975244 (C4 (Web Text))

Max Activating Token Index: 268

Text #15

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2732. Min Act: -0.0000

Data Index: 792444 (C4 (Web Text))

Max Activating Token Index: 909

Text #16

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.3046. Min Act: -0.0000

Data Index: 757093 (C4 (Web Text))

Max Activating Token Index: 387

Text #17

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2508. Min Act: -0.0000

Data Index: 1520548 (Python Code)

Max Activating Token Index: 578

Text #18

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.2586. Min Act: -0.0000

Data Index: 1539257 (Python Code)

Max Activating Token Index: 277

Text #19

Max Range: 2.4135. Min Range: -2.4135

Max Act: 1.1252. Min Act: -0.0000

Data Index: 965147 (C4 (Web Text))

Max Activating Token Index: 318