Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3078 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 2.6158. Min Range: -2.6158

Max Act: 2.6158. Min Act: -0.0000

Data Index: 447786 (C4 (Web Text))

Max Activating Token Index: 45

Text #1

Max Range: 2.6158. Min Range: -2.6158

Max Act: 2.4549. Min Act: -0.0000

Data Index: 911944 (C4 (Web Text))

Max Activating Token Index: 239

Text #2

Max Range: 2.6158. Min Range: -2.6158

Max Act: 2.4498. Min Act: -0.0000

Data Index: 323950 (C4 (Web Text))

Max Activating Token Index: 991

Text #3

Max Range: 2.6158. Min Range: -2.6158

Max Act: 2.3966. Min Act: -0.0000

Data Index: 220605 (C4 (Web Text))

Max Activating Token Index: 601

Text #4

Max Range: 2.6158. Min Range: -2.6158

Max Act: 2.1567. Min Act: -0.0000

Data Index: 225716 (C4 (Web Text))

Max Activating Token Index: 278

Text #5

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.9123. Min Act: -0.0000

Data Index: 1541082 (Python Code)

Max Activating Token Index: 967

Text #6

Max Range: 2.6158. Min Range: -2.6158

Max Act: 2.0555. Min Act: -0.0000

Data Index: 580574 (C4 (Web Text))

Max Activating Token Index: 646

Text #7

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.9854. Min Act: -0.0000

Data Index: 38038 (C4 (Web Text))

Max Activating Token Index: 642

Text #8

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.9704. Min Act: -0.0000

Data Index: 311722 (C4 (Web Text))

Max Activating Token Index: 336

Text #9

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.9705. Min Act: -0.0000

Data Index: 838616 (C4 (Web Text))

Max Activating Token Index: 581

Text #10

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.8255. Min Act: -0.0000

Data Index: 413339 (C4 (Web Text))

Max Activating Token Index: 524

Text #11

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.8483. Min Act: -0.0000

Data Index: 1532223 (Python Code)

Max Activating Token Index: 431

Text #12

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.7824. Min Act: -0.0000

Data Index: 699958 (C4 (Web Text))

Max Activating Token Index: 214

Text #13

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.7224. Min Act: -0.0000

Data Index: 186259 (C4 (Web Text))

Max Activating Token Index: 565

Text #14

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.6316. Min Act: -0.0000

Data Index: 1502735 (Python Code)

Max Activating Token Index: 559

Text #15

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.8130. Min Act: -0.0000

Data Index: 350507 (C4 (Web Text))

Max Activating Token Index: 527

Text #16

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.6647. Min Act: -0.0000

Data Index: 1650142 (Python Code)

Max Activating Token Index: 483

Text #17

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.7551. Min Act: -0.0000

Data Index: 498970 (C4 (Web Text))

Max Activating Token Index: 644

Text #18

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.8513. Min Act: -0.0000

Data Index: 1499505 (Python Code)

Max Activating Token Index: 91

Text #19

Max Range: 2.6158. Min Range: -2.6158

Max Act: 1.7327. Min Act: -0.0000

Data Index: 1279680 (C4 (Web Text))

Max Activating Token Index: 412