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

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

Neuron 2618 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5992. Min Act: -0.0001

Data Index: 991568 (C4 (Web Text))

Max Activating Token Index: 648

Text #1

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5613. Min Act: -0.0001

Data Index: 1338386 (C4 (Web Text))

Max Activating Token Index: 181

Text #2

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5402. Min Act: -0.0001

Data Index: 637338 (C4 (Web Text))

Max Activating Token Index: 217

Text #3

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5538. Min Act: -0.0000

Data Index: 442885 (C4 (Web Text))

Max Activating Token Index: 190

Text #4

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5484. Min Act: -0.0001

Data Index: 1601805 (Python Code)

Max Activating Token Index: 381

Text #5

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5125. Min Act: -0.0001

Data Index: 740558 (C4 (Web Text))

Max Activating Token Index: 622

Text #6

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5422. Min Act: -0.0001

Data Index: 452373 (C4 (Web Text))

Max Activating Token Index: 519

Text #7

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5408. Min Act: -0.0001

Data Index: 498993 (C4 (Web Text))

Max Activating Token Index: 425

Text #8

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5228. Min Act: -0.0001

Data Index: 1629026 (Python Code)

Max Activating Token Index: 360

Text #9

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5196. Min Act: -0.0001

Data Index: 645215 (C4 (Web Text))

Max Activating Token Index: 765

Text #10

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5733. Min Act: -0.0001

Data Index: 160265 (C4 (Web Text))

Max Activating Token Index: 752

Text #11

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5398. Min Act: -0.0001

Data Index: 1608426 (Python Code)

Max Activating Token Index: 250

Text #12

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5324. Min Act: -0.0001

Data Index: 1584714 (Python Code)

Max Activating Token Index: 485

Text #13

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5289. Min Act: -0.0001

Data Index: 651182 (C4 (Web Text))

Max Activating Token Index: 738

Text #14

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5451. Min Act: -0.0001

Data Index: 1567860 (Python Code)

Max Activating Token Index: 657

Text #15

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5342. Min Act: -0.0001

Data Index: 1324479 (C4 (Web Text))

Max Activating Token Index: 729

Text #16

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5317. Min Act: -0.0001

Data Index: 1287147 (C4 (Web Text))

Max Activating Token Index: 126

Text #17

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5321. Min Act: -0.0001

Data Index: 1651921 (Python Code)

Max Activating Token Index: 448

Text #18

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5141. Min Act: -0.0001

Data Index: 588271 (C4 (Web Text))

Max Activating Token Index: 71

Text #19

Max Range: 0.5992. Min Range: -0.5992

Max Act: 0.5487. Min Act: -0.0001

Data Index: 771812 (C4 (Web Text))

Max Activating Token Index: 259