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

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

Neuron 4722 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.7891. Min Act: -0.0002

Data Index: 183273 (C4 (Web Text))

Max Activating Token Index: 180

Text #1

Max Range: 3.2227. Min Range: -3.2227

Max Act: 3.2227. Min Act: -0.0002

Data Index: 452169 (C4 (Web Text))

Max Activating Token Index: 971

Text #2

Max Range: 3.2227. Min Range: -3.2227

Max Act: 3.0738. Min Act: -0.0002

Data Index: 670619 (C4 (Web Text))

Max Activating Token Index: 355

Text #3

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.7444. Min Act: -0.0002

Data Index: 1202737 (C4 (Web Text))

Max Activating Token Index: 231

Text #4

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.8251. Min Act: -0.0002

Data Index: 93640 (C4 (Web Text))

Max Activating Token Index: 160

Text #5

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.8921. Min Act: -0.0002

Data Index: 1145613 (C4 (Web Text))

Max Activating Token Index: 199

Text #6

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.8009. Min Act: -0.0002

Data Index: 406691 (C4 (Web Text))

Max Activating Token Index: 255

Text #7

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.8787. Min Act: -0.0002

Data Index: 657189 (C4 (Web Text))

Max Activating Token Index: 98

Text #8

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.5135. Min Act: -0.0002

Data Index: 859261 (C4 (Web Text))

Max Activating Token Index: 637

Text #9

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.7312. Min Act: -0.0002

Data Index: 1128281 (C4 (Web Text))

Max Activating Token Index: 355

Text #10

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.6388. Min Act: -0.0002

Data Index: 798508 (C4 (Web Text))

Max Activating Token Index: 145

Text #11

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.6388. Min Act: -0.0002

Data Index: 321315 (C4 (Web Text))

Max Activating Token Index: 422

Text #12

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.4656. Min Act: -0.0002

Data Index: 1324126 (C4 (Web Text))

Max Activating Token Index: 156

Text #13

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.6242. Min Act: -0.0002

Data Index: 1038483 (C4 (Web Text))

Max Activating Token Index: 545

Text #14

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.6073. Min Act: -0.0002

Data Index: 515359 (C4 (Web Text))

Max Activating Token Index: 480

Text #15

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.6759. Min Act: -0.0002

Data Index: 383881 (C4 (Web Text))

Max Activating Token Index: 277

Text #16

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.5025. Min Act: -0.0002

Data Index: 578996 (C4 (Web Text))

Max Activating Token Index: 589

Text #17

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.4786. Min Act: -0.0002

Data Index: 1177802 (C4 (Web Text))

Max Activating Token Index: 386

Text #18

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.6138. Min Act: -0.0002

Data Index: 646631 (C4 (Web Text))

Max Activating Token Index: 107

Text #19

Max Range: 3.2227. Min Range: -3.2227

Max Act: 2.6619. Min Act: -0.0002

Data Index: 290094 (C4 (Web Text))

Max Activating Token Index: 1010