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

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

Neuron 133 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7353. Min Range: -3.7353

Max Act: 3.7353. Min Act: -0.0000

Data Index: 476433 (C4 (Web Text))

Max Activating Token Index: 269

Text #1

Max Range: 3.7353. Min Range: -3.7353

Max Act: 3.7292. Min Act: -0.0000

Data Index: 510870 (C4 (Web Text))

Max Activating Token Index: 143

Text #2

Max Range: 3.7353. Min Range: -3.7353

Max Act: 3.5074. Min Act: -0.0000

Data Index: 1323077 (C4 (Web Text))

Max Activating Token Index: 164

Text #3

Max Range: 3.7353. Min Range: -3.7353

Max Act: 3.6408. Min Act: -0.0000

Data Index: 142646 (C4 (Web Text))

Max Activating Token Index: 720

Text #4

Max Range: 3.7353. Min Range: -3.7353

Max Act: 3.1165. Min Act: -0.0000

Data Index: 584819 (C4 (Web Text))

Max Activating Token Index: 732

Text #5

Max Range: 3.7353. Min Range: -3.7353

Max Act: 3.1751. Min Act: -0.0000

Data Index: 711715 (C4 (Web Text))

Max Activating Token Index: 761

Text #6

Max Range: 3.7353. Min Range: -3.7353

Max Act: 3.1729. Min Act: -0.0000

Data Index: 664683 (C4 (Web Text))

Max Activating Token Index: 858

Text #7

Max Range: 3.7353. Min Range: -3.7353

Max Act: 3.0463. Min Act: -0.0000

Data Index: 1186251 (C4 (Web Text))

Max Activating Token Index: 446

Text #8

Max Range: 3.7353. Min Range: -3.7353

Max Act: 3.0342. Min Act: -0.0000

Data Index: 450264 (C4 (Web Text))

Max Activating Token Index: 428

Text #9

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.9500. Min Act: -0.0000

Data Index: 1200304 (C4 (Web Text))

Max Activating Token Index: 277

Text #10

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.9347. Min Act: -0.0000

Data Index: 173414 (C4 (Web Text))

Max Activating Token Index: 579

Text #11

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.9181. Min Act: -0.0000

Data Index: 933960 (C4 (Web Text))

Max Activating Token Index: 484

Text #12

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.9241. Min Act: -0.0000

Data Index: 384623 (C4 (Web Text))

Max Activating Token Index: 496

Text #13

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.9126. Min Act: -0.0000

Data Index: 287628 (C4 (Web Text))

Max Activating Token Index: 995

Text #14

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.8472. Min Act: -0.0000

Data Index: 920835 (C4 (Web Text))

Max Activating Token Index: 680

Text #15

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.8751. Min Act: -0.0000

Data Index: 1331515 (C4 (Web Text))

Max Activating Token Index: 497

Text #16

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.7178. Min Act: -0.0000

Data Index: 317144 (C4 (Web Text))

Max Activating Token Index: 882

Text #17

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.6946. Min Act: -0.0000

Data Index: 1551868 (Python Code)

Max Activating Token Index: 22

Text #18

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.7044. Min Act: -0.0000

Data Index: 868339 (C4 (Web Text))

Max Activating Token Index: 828

Text #19

Max Range: 3.7353. Min Range: -3.7353

Max Act: 2.6998. Min Act: -0.0000

Data Index: 999249 (C4 (Web Text))

Max Activating Token Index: 465