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

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

Neuron 3911 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: 0.5538. Min Range: -0.5538

Max Act: 0.5538. Min Act: -0.0000

Data Index: 36320 (C4 (Web Text))

Max Activating Token Index: 718

Text #1

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.5060. Min Act: -0.0000

Data Index: 894409 (C4 (Web Text))

Max Activating Token Index: 95

Text #2

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.4359. Min Act: -0.0000

Data Index: 813435 (C4 (Web Text))

Max Activating Token Index: 837

Text #3

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.4338. Min Act: -0.0000

Data Index: 413241 (C4 (Web Text))

Max Activating Token Index: 385

Text #4

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.4452. Min Act: -0.0000

Data Index: 1111716 (C4 (Web Text))

Max Activating Token Index: 161

Text #5

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.4454. Min Act: -0.0000

Data Index: 137403 (C4 (Web Text))

Max Activating Token Index: 690

Text #6

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3973. Min Act: -0.0000

Data Index: 1174334 (C4 (Web Text))

Max Activating Token Index: 265

Text #7

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3873. Min Act: -0.0000

Data Index: 831307 (C4 (Web Text))

Max Activating Token Index: 679

Text #8

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3744. Min Act: -0.0000

Data Index: 629392 (C4 (Web Text))

Max Activating Token Index: 990

Text #9

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3370. Min Act: -0.0000

Data Index: 1298250 (C4 (Web Text))

Max Activating Token Index: 597

Text #10

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3527. Min Act: -0.0000

Data Index: 694798 (C4 (Web Text))

Max Activating Token Index: 826

Text #11

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3628. Min Act: -0.0000

Data Index: 145528 (C4 (Web Text))

Max Activating Token Index: 976

Text #12

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3312. Min Act: -0.0000

Data Index: 1469317 (Python Code)

Max Activating Token Index: 31

Text #13

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3207. Min Act: -0.0000

Data Index: 254089 (C4 (Web Text))

Max Activating Token Index: 222

Text #14

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3087. Min Act: -0.0000

Data Index: 1121667 (C4 (Web Text))

Max Activating Token Index: 160

Text #15

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3101. Min Act: -0.0000

Data Index: 59080 (C4 (Web Text))

Max Activating Token Index: 912

Text #16

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3141. Min Act: -0.0000

Data Index: 313104 (C4 (Web Text))

Max Activating Token Index: 902

Text #17

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3143. Min Act: -0.0000

Data Index: 1234762 (C4 (Web Text))

Max Activating Token Index: 530

Text #18

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.2891. Min Act: -0.0000

Data Index: 570913 (C4 (Web Text))

Max Activating Token Index: 842

Text #19

Max Range: 0.5538. Min Range: -0.5538

Max Act: 0.3163. Min Act: -0.0000

Data Index: 1320321 (C4 (Web Text))

Max Activating Token Index: 145