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

Dataset: The Pile

Neuron 256 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.2120. Min Act: -0.0000

Data Index: 924144 (The Pile)

Max Activating Token Index: 828

Text #1

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.2602. Min Act: -0.0000

Data Index: 593815 (The Pile)

Max Activating Token Index: 914

Text #2

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.2703. Min Act: -0.0000

Data Index: 87189 (The Pile)

Max Activating Token Index: 259

Text #3

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.1855. Min Act: -0.0000

Data Index: 19650 (The Pile)

Max Activating Token Index: 188

Text #4

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.1070. Min Act: -0.0000

Data Index: 1259447 (The Pile)

Max Activating Token Index: 966

Text #5

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.1467. Min Act: -0.0000

Data Index: 919698 (The Pile)

Max Activating Token Index: 672

Text #6

Max Range: 1.2703. Min Range: -1.2703

Max Act: 0.9561. Min Act: -0.0000

Data Index: 1473378 (The Pile)

Max Activating Token Index: 898

Text #7

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.0698. Min Act: -0.0000

Data Index: 543856 (The Pile)

Max Activating Token Index: 447

Text #8

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.0225. Min Act: -0.0000

Data Index: 571856 (The Pile)

Max Activating Token Index: 762

Text #9

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.0007. Min Act: -0.0000

Data Index: 1040249 (The Pile)

Max Activating Token Index: 768

Text #10

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.0962. Min Act: -0.0000

Data Index: 871936 (The Pile)

Max Activating Token Index: 279

Text #11

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.1210. Min Act: -0.0000

Data Index: 617066 (The Pile)

Max Activating Token Index: 861

Text #12

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.0415. Min Act: -0.0000

Data Index: 1089582 (The Pile)

Max Activating Token Index: 741

Text #13

Max Range: 1.2703. Min Range: -1.2703

Max Act: 0.9632. Min Act: -0.0000

Data Index: 1165004 (The Pile)

Max Activating Token Index: 429

Text #14

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.0089. Min Act: -0.0000

Data Index: 1336766 (The Pile)

Max Activating Token Index: 471

Text #15

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.0093. Min Act: -0.0000

Data Index: 803689 (The Pile)

Max Activating Token Index: 976

Text #16

Max Range: 1.2703. Min Range: -1.2703

Max Act: 0.9841. Min Act: -0.0000

Data Index: 178643 (The Pile)

Max Activating Token Index: 785

Text #17

Max Range: 1.2703. Min Range: -1.2703

Max Act: 0.9707. Min Act: -0.0000

Data Index: 1380295 (The Pile)

Max Activating Token Index: 418

Text #18

Max Range: 1.2703. Min Range: -1.2703

Max Act: 0.9942. Min Act: -0.0000

Data Index: 1135005 (The Pile)

Max Activating Token Index: 397

Text #19

Max Range: 1.2703. Min Range: -1.2703

Max Act: 1.0162. Min Act: -0.0000

Data Index: 123352 (The Pile)

Max Activating Token Index: 478