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

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

Neuron 547 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2818. Min Act: -0.0001

Data Index: 191497 (C4 (Web Text))

Max Activating Token Index: 329

Text #1

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2454. Min Act: -0.0001

Data Index: 238189 (C4 (Web Text))

Max Activating Token Index: 70

Text #2

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2368. Min Act: -0.0001

Data Index: 144812 (C4 (Web Text))

Max Activating Token Index: 572

Text #3

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2329. Min Act: -0.0001

Data Index: 444545 (C4 (Web Text))

Max Activating Token Index: 179

Text #4

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2300. Min Act: -0.0001

Data Index: 242277 (C4 (Web Text))

Max Activating Token Index: 50

Text #5

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2263. Min Act: -0.0001

Data Index: 547962 (C4 (Web Text))

Max Activating Token Index: 67

Text #6

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2150. Min Act: -0.0001

Data Index: 1234564 (C4 (Web Text))

Max Activating Token Index: 37

Text #7

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2367. Min Act: -0.0001

Data Index: 1079354 (C4 (Web Text))

Max Activating Token Index: 655

Text #8

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2122. Min Act: -0.0001

Data Index: 459098 (C4 (Web Text))

Max Activating Token Index: 440

Text #9

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2017. Min Act: -0.0001

Data Index: 354198 (C4 (Web Text))

Max Activating Token Index: 570

Text #10

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2115. Min Act: -0.0001

Data Index: 222297 (C4 (Web Text))

Max Activating Token Index: 548

Text #11

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.1879. Min Act: -0.0001

Data Index: 1114200 (C4 (Web Text))

Max Activating Token Index: 688

Text #12

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2011. Min Act: -0.0001

Data Index: 1349391 (C4 (Web Text))

Max Activating Token Index: 20

Text #13

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.2046. Min Act: -0.0001

Data Index: 617785 (C4 (Web Text))

Max Activating Token Index: 309

Text #14

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.1795. Min Act: -0.0001

Data Index: 811407 (C4 (Web Text))

Max Activating Token Index: 338

Text #15

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.1981. Min Act: -0.0001

Data Index: 683456 (C4 (Web Text))

Max Activating Token Index: 507

Text #16

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.1978. Min Act: -0.0001

Data Index: 272941 (C4 (Web Text))

Max Activating Token Index: 446

Text #17

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.1823. Min Act: -0.0001

Data Index: 753460 (C4 (Web Text))

Max Activating Token Index: 543

Text #18

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.1864. Min Act: -0.0001

Data Index: 239440 (C4 (Web Text))

Max Activating Token Index: 428

Text #19

Max Range: 0.2818. Min Range: -0.2818

Max Act: 0.1856. Min Act: -0.0001

Data Index: 653819 (C4 (Web Text))

Max Activating Token Index: 918