Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3973 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.5017. Min Act: -0.0001

Data Index: 1651312 (Python Code)

Max Activating Token Index: 381

Text #1

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.4588. Min Act: -0.0001

Data Index: 1481673 (Python Code)

Max Activating Token Index: 130

Text #2

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3967. Min Act: -0.0000

Data Index: 1460196 (Python Code)

Max Activating Token Index: 824

Text #3

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.4178. Min Act: -0.0001

Data Index: 1382121 (Python Code)

Max Activating Token Index: 1015

Text #4

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.4027. Min Act: -0.0000

Data Index: 1516978 (Python Code)

Max Activating Token Index: 712

Text #5

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.4019. Min Act: -0.0000

Data Index: 1266408 (C4 (Web Text))

Max Activating Token Index: 594

Text #6

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3679. Min Act: -0.0000

Data Index: 1248382 (C4 (Web Text))

Max Activating Token Index: 876

Text #7

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3494. Min Act: -0.0000

Data Index: 1024357 (C4 (Web Text))

Max Activating Token Index: 462

Text #8

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3608. Min Act: -0.0000

Data Index: 94738 (C4 (Web Text))

Max Activating Token Index: 202

Text #9

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3339. Min Act: -0.0000

Data Index: 622524 (C4 (Web Text))

Max Activating Token Index: 966

Text #10

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3579. Min Act: -0.0000

Data Index: 682073 (C4 (Web Text))

Max Activating Token Index: 456

Text #11

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3422. Min Act: -0.0001

Data Index: 1531706 (Python Code)

Max Activating Token Index: 37

Text #12

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3248. Min Act: -0.0001

Data Index: 1416170 (Python Code)

Max Activating Token Index: 210

Text #13

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3308. Min Act: -0.0000

Data Index: 153673 (C4 (Web Text))

Max Activating Token Index: 416

Text #14

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3480. Min Act: -0.0000

Data Index: 1312566 (C4 (Web Text))

Max Activating Token Index: 512

Text #15

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3343. Min Act: -0.0001

Data Index: 761313 (C4 (Web Text))

Max Activating Token Index: 345

Text #16

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3441. Min Act: -0.0000

Data Index: 1040138 (C4 (Web Text))

Max Activating Token Index: 999

Text #17

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3306. Min Act: -0.0000

Data Index: 663017 (C4 (Web Text))

Max Activating Token Index: 311

Text #18

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3182. Min Act: -0.0000

Data Index: 698416 (C4 (Web Text))

Max Activating Token Index: 76

Text #19

Max Range: 0.5017. Min Range: -0.5017

Max Act: 0.3285. Min Act: -0.0000

Data Index: 174222 (C4 (Web Text))

Max Activating Token Index: 129