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

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

Neuron 533 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.8331. Min Act: -0.0001

Data Index: 24253 (C4 (Web Text))

Max Activating Token Index: 424

Text #1

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.8089. Min Act: -0.0000

Data Index: 695700 (C4 (Web Text))

Max Activating Token Index: 484

Text #2

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.8320. Min Act: 0.0001

Data Index: 642799 (C4 (Web Text))

Max Activating Token Index: 256

Text #3

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.7394. Min Act: -0.0000

Data Index: 1214944 (C4 (Web Text))

Max Activating Token Index: 539

Text #4

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.7052. Min Act: -0.0001

Data Index: 1496188 (Python Code)

Max Activating Token Index: 251

Text #5

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.7089. Min Act: -0.0000

Data Index: 957340 (C4 (Web Text))

Max Activating Token Index: 231

Text #6

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.6381. Min Act: -0.0000

Data Index: 1336024 (C4 (Web Text))

Max Activating Token Index: 158

Text #7

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.6047. Min Act: -0.0000

Data Index: 1350636 (C4 (Web Text))

Max Activating Token Index: 564

Text #8

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.6200. Min Act: -0.0000

Data Index: 1092515 (C4 (Web Text))

Max Activating Token Index: 539

Text #9

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.6354. Min Act: -0.0000

Data Index: 522861 (C4 (Web Text))

Max Activating Token Index: 18

Text #10

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.5905. Min Act: -0.0001

Data Index: 1506780 (Python Code)

Max Activating Token Index: 468

Text #11

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.5855. Min Act: -0.0000

Data Index: 641925 (C4 (Web Text))

Max Activating Token Index: 112

Text #12

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.6018. Min Act: -0.0000

Data Index: 1156971 (C4 (Web Text))

Max Activating Token Index: 646

Text #13

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.5888. Min Act: -0.0001

Data Index: 1199761 (C4 (Web Text))

Max Activating Token Index: 716

Text #14

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.5369. Min Act: -0.0000

Data Index: 217386 (C4 (Web Text))

Max Activating Token Index: 965

Text #15

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.5456. Min Act: -0.0000

Data Index: 290036 (C4 (Web Text))

Max Activating Token Index: 450

Text #16

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.4922. Min Act: -0.0000

Data Index: 231358 (C4 (Web Text))

Max Activating Token Index: 278

Text #17

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.5556. Min Act: -0.0001

Data Index: 1531956 (Python Code)

Max Activating Token Index: 170

Text #18

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.5504. Min Act: -0.0000

Data Index: 480034 (C4 (Web Text))

Max Activating Token Index: 620

Text #19

Max Range: 0.8331. Min Range: -0.8331

Max Act: 0.5509. Min Act: -0.0000

Data Index: 1093593 (C4 (Web Text))

Max Activating Token Index: 267