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

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

Neuron 2550 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: 2.0003. Min Range: -2.0003

Max Act: 2.0003. Min Act: -0.0000

Data Index: 650176 (C4 (Web Text))

Max Activating Token Index: 99

Text #1

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.7079. Min Act: 0.0000

Data Index: 1073813 (C4 (Web Text))

Max Activating Token Index: 198

Text #2

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.4140. Min Act: 0.0000

Data Index: 1026660 (C4 (Web Text))

Max Activating Token Index: 395

Text #3

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.2246. Min Act: -0.0000

Data Index: 271184 (C4 (Web Text))

Max Activating Token Index: 241

Text #4

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.2866. Min Act: -0.0000

Data Index: 262639 (C4 (Web Text))

Max Activating Token Index: 369

Text #5

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.2972. Min Act: -0.0000

Data Index: 899105 (C4 (Web Text))

Max Activating Token Index: 294

Text #6

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.0481. Min Act: -0.0000

Data Index: 632261 (C4 (Web Text))

Max Activating Token Index: 267

Text #7

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.0836. Min Act: 0.0000

Data Index: 160473 (C4 (Web Text))

Max Activating Token Index: 195

Text #8

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.0793. Min Act: -0.0000

Data Index: 1273649 (C4 (Web Text))

Max Activating Token Index: 570

Text #9

Max Range: 2.0003. Min Range: -2.0003

Max Act: 0.9835. Min Act: -0.0000

Data Index: 5710 (C4 (Web Text))

Max Activating Token Index: 303

Text #10

Max Range: 2.0003. Min Range: -2.0003

Max Act: 0.9653. Min Act: -0.0000

Data Index: 831579 (C4 (Web Text))

Max Activating Token Index: 830

Text #11

Max Range: 2.0003. Min Range: -2.0003

Max Act: 0.9542. Min Act: -0.0000

Data Index: 854498 (C4 (Web Text))

Max Activating Token Index: 815

Text #12

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.0123. Min Act: -0.0000

Data Index: 89455 (C4 (Web Text))

Max Activating Token Index: 966

Text #13

Max Range: 2.0003. Min Range: -2.0003

Max Act: 0.9503. Min Act: -0.0000

Data Index: 751864 (C4 (Web Text))

Max Activating Token Index: 321

Text #14

Max Range: 2.0003. Min Range: -2.0003

Max Act: 0.9989. Min Act: 0.0000

Data Index: 614403 (C4 (Web Text))

Max Activating Token Index: 512

Text #15

Max Range: 2.0003. Min Range: -2.0003

Max Act: 0.9525. Min Act: -0.0000

Data Index: 49947 (C4 (Web Text))

Max Activating Token Index: 711

Text #16

Max Range: 2.0003. Min Range: -2.0003

Max Act: 1.0178. Min Act: -0.0000

Data Index: 170817 (C4 (Web Text))

Max Activating Token Index: 387

Text #17

Max Range: 2.0003. Min Range: -2.0003

Max Act: 0.9843. Min Act: 0.0000

Data Index: 1096515 (C4 (Web Text))

Max Activating Token Index: 866

Text #18

Max Range: 2.0003. Min Range: -2.0003

Max Act: 0.9516. Min Act: 0.0000

Data Index: 99297 (C4 (Web Text))

Max Activating Token Index: 866

Text #19

Max Range: 2.0003. Min Range: -2.0003

Max Act: 0.9148. Min Act: -0.0000

Data Index: 541654 (C4 (Web Text))

Max Activating Token Index: 525