Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 773 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.5573. Min Act: -0.0001

Data Index: 104387 (C4 (Web Text))

Max Activating Token Index: 58

Text #1

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.5518. Min Act: -0.0001

Data Index: 908373 (C4 (Web Text))

Max Activating Token Index: 71

Text #2

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.5344. Min Act: -0.0001

Data Index: 1073249 (C4 (Web Text))

Max Activating Token Index: 773

Text #3

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.5131. Min Act: -0.0001

Data Index: 1168871 (C4 (Web Text))

Max Activating Token Index: 147

Text #4

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.5060. Min Act: -0.0001

Data Index: 480902 (C4 (Web Text))

Max Activating Token Index: 973

Text #5

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.5231. Min Act: -0.0001

Data Index: 1612294 (Python Code)

Max Activating Token Index: 144

Text #6

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4862. Min Act: -0.0001

Data Index: 862389 (C4 (Web Text))

Max Activating Token Index: 917

Text #7

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.5303. Min Act: -0.0001

Data Index: 77233 (C4 (Web Text))

Max Activating Token Index: 452

Text #8

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.5313. Min Act: -0.0001

Data Index: 986772 (C4 (Web Text))

Max Activating Token Index: 941

Text #9

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4992. Min Act: -0.0001

Data Index: 916122 (C4 (Web Text))

Max Activating Token Index: 1021

Text #10

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4791. Min Act: -0.0001

Data Index: 1045923 (C4 (Web Text))

Max Activating Token Index: 378

Text #11

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4953. Min Act: -0.0001

Data Index: 426617 (C4 (Web Text))

Max Activating Token Index: 1000

Text #12

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4949. Min Act: -0.0001

Data Index: 1262797 (C4 (Web Text))

Max Activating Token Index: 773

Text #13

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4915. Min Act: -0.0001

Data Index: 473075 (C4 (Web Text))

Max Activating Token Index: 1022

Text #14

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4487. Min Act: -0.0001

Data Index: 652170 (C4 (Web Text))

Max Activating Token Index: 735

Text #15

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4682. Min Act: -0.0001

Data Index: 1133340 (C4 (Web Text))

Max Activating Token Index: 520

Text #16

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4360. Min Act: -0.0001

Data Index: 181792 (C4 (Web Text))

Max Activating Token Index: 157

Text #17

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4832. Min Act: -0.0001

Data Index: 1629480 (Python Code)

Max Activating Token Index: 818

Text #18

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4518. Min Act: -0.0001

Data Index: 1328577 (C4 (Web Text))

Max Activating Token Index: 406

Text #19

Max Range: 0.5573. Min Range: -0.5573

Max Act: 0.4792. Min Act: -0.0001

Data Index: 792597 (C4 (Web Text))

Max Activating Token Index: 52