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

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

Neuron 2350 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.9687. Min Act: -0.0001

Data Index: 426403 (C4 (Web Text))

Max Activating Token Index: 38

Text #1

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.7776. Min Act: -0.0001

Data Index: 1459237 (Python Code)

Max Activating Token Index: 777

Text #2

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.7799. Min Act: -0.0001

Data Index: 1522514 (Python Code)

Max Activating Token Index: 642

Text #3

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.8071. Min Act: -0.0001

Data Index: 1392133 (Python Code)

Max Activating Token Index: 840

Text #4

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.7481. Min Act: -0.0001

Data Index: 1655367 (Python Code)

Max Activating Token Index: 662

Text #5

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.7663. Min Act: -0.0001

Data Index: 1371385 (Python Code)

Max Activating Token Index: 919

Text #6

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.7032. Min Act: -0.0001

Data Index: 1590432 (Python Code)

Max Activating Token Index: 691

Text #7

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.7296. Min Act: -0.0001

Data Index: 1417707 (Python Code)

Max Activating Token Index: 288

Text #8

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.7158. Min Act: -0.0000

Data Index: 591358 (C4 (Web Text))

Max Activating Token Index: 46

Text #9

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6660. Min Act: -0.0001

Data Index: 1461654 (Python Code)

Max Activating Token Index: 254

Text #10

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6822. Min Act: -0.0000

Data Index: 1049150 (C4 (Web Text))

Max Activating Token Index: 163

Text #11

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6606. Min Act: -0.0001

Data Index: 1514897 (Python Code)

Max Activating Token Index: 19

Text #12

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6620. Min Act: -0.0001

Data Index: 1460305 (Python Code)

Max Activating Token Index: 371

Text #13

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6159. Min Act: -0.0001

Data Index: 1519011 (Python Code)

Max Activating Token Index: 528

Text #14

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6440. Min Act: -0.0000

Data Index: 920388 (C4 (Web Text))

Max Activating Token Index: 645

Text #15

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6197. Min Act: -0.0001

Data Index: 1552708 (Python Code)

Max Activating Token Index: 901

Text #16

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6312. Min Act: -0.0001

Data Index: 1529147 (Python Code)

Max Activating Token Index: 400

Text #17

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6503. Min Act: -0.0001

Data Index: 1544771 (Python Code)

Max Activating Token Index: 475

Text #18

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6044. Min Act: -0.0001

Data Index: 1582392 (Python Code)

Max Activating Token Index: 41

Text #19

Max Range: 0.9687. Min Range: -0.9687

Max Act: 0.6417. Min Act: -0.0000

Data Index: 1211387 (C4 (Web Text))

Max Activating Token Index: 446