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

Dataset: The Pile

Neuron 3150 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.7762. Min Range: -8.7762

Max Act: 8.7020. Min Act: -0.0000

Data Index: 1572440 (The Pile)

Max Activating Token Index: 406

Text #1

Max Range: 8.7762. Min Range: -8.7762

Max Act: 8.5150. Min Act: -0.0000

Data Index: 689789 (The Pile)

Max Activating Token Index: 383

Text #2

Max Range: 8.7762. Min Range: -8.7762

Max Act: 8.7762. Min Act: -0.0000

Data Index: 13741 (The Pile)

Max Activating Token Index: 873

Text #3

Max Range: 8.7762. Min Range: -8.7762

Max Act: 8.0709. Min Act: -0.0000

Data Index: 1040452 (The Pile)

Max Activating Token Index: 210

Text #4

Max Range: 8.7762. Min Range: -8.7762

Max Act: 8.2429. Min Act: -0.0000

Data Index: 357801 (The Pile)

Max Activating Token Index: 244

Text #5

Max Range: 8.7762. Min Range: -8.7762

Max Act: 8.0663. Min Act: -0.0000

Data Index: 1476038 (The Pile)

Max Activating Token Index: 99

Text #6

Max Range: 8.7762. Min Range: -8.7762

Max Act: 7.6097. Min Act: -0.0000

Data Index: 1764098 (The Pile)

Max Activating Token Index: 567

Text #7

Max Range: 8.7762. Min Range: -8.7762

Max Act: 7.5437. Min Act: -0.0000

Data Index: 1378700 (The Pile)

Max Activating Token Index: 187

Text #8

Max Range: 8.7762. Min Range: -8.7762

Max Act: 6.8902. Min Act: -0.0000

Data Index: 88351 (The Pile)

Max Activating Token Index: 430

Text #9

Max Range: 8.7762. Min Range: -8.7762

Max Act: 6.9642. Min Act: -0.0000

Data Index: 323849 (The Pile)

Max Activating Token Index: 622

Text #10

Max Range: 8.7762. Min Range: -8.7762

Max Act: 6.7015. Min Act: -0.0000

Data Index: 401495 (The Pile)

Max Activating Token Index: 938

Text #11

Max Range: 8.7762. Min Range: -8.7762

Max Act: 5.3755. Min Act: -0.0000

Data Index: 1545290 (The Pile)

Max Activating Token Index: 841

Text #12

Max Range: 8.7762. Min Range: -8.7762

Max Act: 5.3043. Min Act: -0.0000

Data Index: 1548345 (The Pile)

Max Activating Token Index: 761

Text #13

Max Range: 8.7762. Min Range: -8.7762

Max Act: 5.2818. Min Act: -0.0000

Data Index: 234662 (The Pile)

Max Activating Token Index: 349

Text #14

Max Range: 8.7762. Min Range: -8.7762

Max Act: 4.9896. Min Act: -0.0000

Data Index: 1043062 (The Pile)

Max Activating Token Index: 489

Text #15

Max Range: 8.7762. Min Range: -8.7762

Max Act: 4.0691. Min Act: -0.0000

Data Index: 615145 (The Pile)

Max Activating Token Index: 913

Text #16

Max Range: 8.7762. Min Range: -8.7762

Max Act: 4.8286. Min Act: -0.0000

Data Index: 1445664 (The Pile)

Max Activating Token Index: 396

Text #17

Max Range: 8.7762. Min Range: -8.7762

Max Act: 4.7946. Min Act: -0.0000

Data Index: 1815191 (The Pile)

Max Activating Token Index: 571

Text #18

Max Range: 8.7762. Min Range: -8.7762

Max Act: 4.5915. Min Act: -0.0000

Data Index: 34058 (The Pile)

Max Activating Token Index: 309

Text #19

Max Range: 8.7762. Min Range: -8.7762

Max Act: 4.7387. Min Act: -0.0000

Data Index: 27446 (The Pile)

Max Activating Token Index: 654