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

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

Neuron 3047 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: 1.2583. Min Range: -1.2583

Max Act: 1.2583. Min Act: -0.0001

Data Index: 245434 (C4 (Web Text))

Max Activating Token Index: 519

Text #1

Max Range: 1.2583. Min Range: -1.2583

Max Act: 1.1625. Min Act: -0.0001

Data Index: 15894 (C4 (Web Text))

Max Activating Token Index: 252

Text #2

Max Range: 1.2583. Min Range: -1.2583

Max Act: 1.2010. Min Act: -0.0001

Data Index: 231820 (C4 (Web Text))

Max Activating Token Index: 847

Text #3

Max Range: 1.2583. Min Range: -1.2583

Max Act: 1.0535. Min Act: -0.0001

Data Index: 820834 (C4 (Web Text))

Max Activating Token Index: 447

Text #4

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.9629. Min Act: -0.0001

Data Index: 1200986 (C4 (Web Text))

Max Activating Token Index: 372

Text #5

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.9788. Min Act: -0.0001

Data Index: 192987 (C4 (Web Text))

Max Activating Token Index: 343

Text #6

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.9702. Min Act: -0.0001

Data Index: 1311634 (C4 (Web Text))

Max Activating Token Index: 576

Text #7

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.9959. Min Act: -0.0001

Data Index: 476372 (C4 (Web Text))

Max Activating Token Index: 653

Text #8

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.9174. Min Act: -0.0001

Data Index: 843879 (C4 (Web Text))

Max Activating Token Index: 569

Text #9

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.8982. Min Act: -0.0001

Data Index: 1456040 (Python Code)

Max Activating Token Index: 413

Text #10

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.8836. Min Act: -0.0001

Data Index: 35171 (C4 (Web Text))

Max Activating Token Index: 507

Text #11

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.9132. Min Act: -0.0001

Data Index: 351228 (C4 (Web Text))

Max Activating Token Index: 200

Text #12

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.8408. Min Act: -0.0001

Data Index: 1409896 (Python Code)

Max Activating Token Index: 958

Text #13

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.8286. Min Act: -0.0001

Data Index: 723266 (C4 (Web Text))

Max Activating Token Index: 84

Text #14

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.8629. Min Act: -0.0001

Data Index: 1388772 (Python Code)

Max Activating Token Index: 28

Text #15

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.8381. Min Act: -0.0001

Data Index: 1385440 (Python Code)

Max Activating Token Index: 394

Text #16

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.8230. Min Act: -0.0001

Data Index: 1054225 (C4 (Web Text))

Max Activating Token Index: 128

Text #17

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.7930. Min Act: -0.0001

Data Index: 1417049 (Python Code)

Max Activating Token Index: 427

Text #18

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.7660. Min Act: -0.0001

Data Index: 1511075 (Python Code)

Max Activating Token Index: 18

Text #19

Max Range: 1.2583. Min Range: -1.2583

Max Act: 0.7930. Min Act: -0.0001

Data Index: 1391651 (Python Code)

Max Activating Token Index: 199