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

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

Neuron 1659 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.6583. Min Act: -0.0001

Data Index: 1234842 (C4 (Web Text))

Max Activating Token Index: 69

Text #1

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.6248. Min Act: -0.0001

Data Index: 1310317 (C4 (Web Text))

Max Activating Token Index: 559

Text #2

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5872. Min Act: -0.0001

Data Index: 993623 (C4 (Web Text))

Max Activating Token Index: 276

Text #3

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5595. Min Act: -0.0001

Data Index: 1195919 (C4 (Web Text))

Max Activating Token Index: 395

Text #4

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5680. Min Act: -0.0001

Data Index: 1288607 (C4 (Web Text))

Max Activating Token Index: 213

Text #5

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5435. Min Act: -0.0001

Data Index: 676057 (C4 (Web Text))

Max Activating Token Index: 54

Text #6

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5429. Min Act: -0.0001

Data Index: 827364 (C4 (Web Text))

Max Activating Token Index: 986

Text #7

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5664. Min Act: -0.0001

Data Index: 756239 (C4 (Web Text))

Max Activating Token Index: 116

Text #8

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5647. Min Act: -0.0001

Data Index: 335070 (C4 (Web Text))

Max Activating Token Index: 46

Text #9

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5625. Min Act: -0.0001

Data Index: 689290 (C4 (Web Text))

Max Activating Token Index: 370

Text #10

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5429. Min Act: -0.0001

Data Index: 620896 (C4 (Web Text))

Max Activating Token Index: 486

Text #11

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5594. Min Act: -0.0001

Data Index: 450319 (C4 (Web Text))

Max Activating Token Index: 300

Text #12

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5563. Min Act: -0.0001

Data Index: 1347 (C4 (Web Text))

Max Activating Token Index: 866

Text #13

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5740. Min Act: -0.0001

Data Index: 854295 (C4 (Web Text))

Max Activating Token Index: 239

Text #14

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5135. Min Act: -0.0001

Data Index: 123011 (C4 (Web Text))

Max Activating Token Index: 434

Text #15

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5072. Min Act: -0.0001

Data Index: 189864 (C4 (Web Text))

Max Activating Token Index: 24

Text #16

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5495. Min Act: -0.0001

Data Index: 886653 (C4 (Web Text))

Max Activating Token Index: 230

Text #17

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.4993. Min Act: -0.0001

Data Index: 1459730 (Python Code)

Max Activating Token Index: 537

Text #18

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.4990. Min Act: -0.0001

Data Index: 494850 (C4 (Web Text))

Max Activating Token Index: 698

Text #19

Max Range: 0.6583. Min Range: -0.6583

Max Act: 0.5005. Min Act: -0.0001

Data Index: 338840 (C4 (Web Text))

Max Activating Token Index: 578