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

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

Neuron 3638 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.5860. Min Act: -0.0001

Data Index: 493936 (C4 (Web Text))

Max Activating Token Index: 16

Text #1

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.5979. Min Act: -0.0000

Data Index: 347287 (C4 (Web Text))

Max Activating Token Index: 31

Text #2

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.5458. Min Act: -0.0000

Data Index: 565476 (C4 (Web Text))

Max Activating Token Index: 585

Text #3

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.5077. Min Act: -0.0000

Data Index: 1170245 (C4 (Web Text))

Max Activating Token Index: 310

Text #4

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.5163. Min Act: -0.0000

Data Index: 1082639 (C4 (Web Text))

Max Activating Token Index: 138

Text #5

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4154. Min Act: -0.0000

Data Index: 497187 (C4 (Web Text))

Max Activating Token Index: 680

Text #6

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4484. Min Act: -0.0001

Data Index: 869481 (C4 (Web Text))

Max Activating Token Index: 76

Text #7

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4528. Min Act: -0.0000

Data Index: 1180256 (C4 (Web Text))

Max Activating Token Index: 72

Text #8

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4081. Min Act: -0.0000

Data Index: 802564 (C4 (Web Text))

Max Activating Token Index: 849

Text #9

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4223. Min Act: -0.0000

Data Index: 464182 (C4 (Web Text))

Max Activating Token Index: 766

Text #10

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.3985. Min Act: -0.0001

Data Index: 345136 (C4 (Web Text))

Max Activating Token Index: 815

Text #11

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.3919. Min Act: -0.0001

Data Index: 164591 (C4 (Web Text))

Max Activating Token Index: 617

Text #12

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4453. Min Act: -0.0001

Data Index: 641552 (C4 (Web Text))

Max Activating Token Index: 788

Text #13

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.3870. Min Act: -0.0000

Data Index: 1316001 (C4 (Web Text))

Max Activating Token Index: 633

Text #14

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4271. Min Act: -0.0000

Data Index: 20850 (C4 (Web Text))

Max Activating Token Index: 607

Text #15

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4440. Min Act: -0.0000

Data Index: 215848 (C4 (Web Text))

Max Activating Token Index: 973

Text #16

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4223. Min Act: -0.0000

Data Index: 276896 (C4 (Web Text))

Max Activating Token Index: 812

Text #17

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4246. Min Act: -0.0000

Data Index: 582820 (C4 (Web Text))

Max Activating Token Index: 843

Text #18

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.4008. Min Act: -0.0001

Data Index: 1330556 (C4 (Web Text))

Max Activating Token Index: 210

Text #19

Max Range: 0.5979. Min Range: -0.5979

Max Act: 0.3915. Min Act: -0.0001

Data Index: 1039032 (C4 (Web Text))

Max Activating Token Index: 566