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

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

Neuron 4851 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.0904. Min Act: -0.0000

Data Index: 1258736 (C4 (Web Text))

Max Activating Token Index: 992

Text #1

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.2778. Min Act: -0.0000

Data Index: 729946 (C4 (Web Text))

Max Activating Token Index: 960

Text #2

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.1130. Min Act: -0.0000

Data Index: 745529 (C4 (Web Text))

Max Activating Token Index: 310

Text #3

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.1960. Min Act: -0.0000

Data Index: 126055 (C4 (Web Text))

Max Activating Token Index: 976

Text #4

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.1934. Min Act: -0.0000

Data Index: 662962 (C4 (Web Text))

Max Activating Token Index: 769

Text #5

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.1620. Min Act: -0.0000

Data Index: 851865 (C4 (Web Text))

Max Activating Token Index: 245

Text #6

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.0285. Min Act: -0.0000

Data Index: 1196349 (C4 (Web Text))

Max Activating Token Index: 391

Text #7

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.8170. Min Act: -0.0000

Data Index: 1259521 (C4 (Web Text))

Max Activating Token Index: 783

Text #8

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.0820. Min Act: -0.0000

Data Index: 672483 (C4 (Web Text))

Max Activating Token Index: 597

Text #9

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.8979. Min Act: -0.0000

Data Index: 887573 (C4 (Web Text))

Max Activating Token Index: 614

Text #10

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.8887. Min Act: -0.0000

Data Index: 469083 (C4 (Web Text))

Max Activating Token Index: 477

Text #11

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.8248. Min Act: -0.0000

Data Index: 1371848 (Python Code)

Max Activating Token Index: 787

Text #12

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.9315. Min Act: -0.0000

Data Index: 1314306 (C4 (Web Text))

Max Activating Token Index: 495

Text #13

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.9333. Min Act: -0.0000

Data Index: 931411 (C4 (Web Text))

Max Activating Token Index: 742

Text #14

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.2867. Min Act: -0.0000

Data Index: 1183814 (C4 (Web Text))

Max Activating Token Index: 614

Text #15

Max Range: 2.2867. Min Range: -2.2867

Max Act: 2.0165. Min Act: -0.0000

Data Index: 735060 (C4 (Web Text))

Max Activating Token Index: 373

Text #16

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.7275. Min Act: -0.0000

Data Index: 1511598 (Python Code)

Max Activating Token Index: 225

Text #17

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.8572. Min Act: -0.0000

Data Index: 963992 (C4 (Web Text))

Max Activating Token Index: 532

Text #18

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.6421. Min Act: -0.0000

Data Index: 1079197 (C4 (Web Text))

Max Activating Token Index: 237

Text #19

Max Range: 2.2867. Min Range: -2.2867

Max Act: 1.8564. Min Act: -0.0000

Data Index: 1622858 (Python Code)

Max Activating Token Index: 836