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

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

Neuron 2820 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: 0.4739. Min Range: -0.4739

Max Act: 0.4739. Min Act: -0.0001

Data Index: 763220 (C4 (Web Text))

Max Activating Token Index: 892

Text #1

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.4147. Min Act: -0.0000

Data Index: 688518 (C4 (Web Text))

Max Activating Token Index: 633

Text #2

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.3301. Min Act: -0.0001

Data Index: 1094550 (C4 (Web Text))

Max Activating Token Index: 457

Text #3

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2933. Min Act: -0.0001

Data Index: 776462 (C4 (Web Text))

Max Activating Token Index: 637

Text #4

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2816. Min Act: -0.0000

Data Index: 190282 (C4 (Web Text))

Max Activating Token Index: 532

Text #5

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2657. Min Act: -0.0000

Data Index: 1609613 (Python Code)

Max Activating Token Index: 80

Text #6

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2708. Min Act: -0.0000

Data Index: 937616 (C4 (Web Text))

Max Activating Token Index: 130

Text #7

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2804. Min Act: -0.0001

Data Index: 1489019 (Python Code)

Max Activating Token Index: 279

Text #8

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2691. Min Act: -0.0001

Data Index: 1486414 (Python Code)

Max Activating Token Index: 101

Text #9

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2561. Min Act: -0.0000

Data Index: 65815 (C4 (Web Text))

Max Activating Token Index: 270

Text #10

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2596. Min Act: -0.0001

Data Index: 298401 (C4 (Web Text))

Max Activating Token Index: 464

Text #11

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2697. Min Act: -0.0000

Data Index: 619628 (C4 (Web Text))

Max Activating Token Index: 759

Text #12

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2583. Min Act: -0.0001

Data Index: 1504699 (Python Code)

Max Activating Token Index: 598

Text #13

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2531. Min Act: -0.0000

Data Index: 1476066 (Python Code)

Max Activating Token Index: 60

Text #14

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2412. Min Act: -0.0000

Data Index: 817839 (C4 (Web Text))

Max Activating Token Index: 331

Text #15

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2473. Min Act: -0.0000

Data Index: 214375 (C4 (Web Text))

Max Activating Token Index: 118

Text #16

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2350. Min Act: -0.0001

Data Index: 1506767 (Python Code)

Max Activating Token Index: 324

Text #17

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2464. Min Act: -0.0000

Data Index: 318724 (C4 (Web Text))

Max Activating Token Index: 416

Text #18

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2448. Min Act: -0.0001

Data Index: 739235 (C4 (Web Text))

Max Activating Token Index: 404

Text #19

Max Range: 0.4739. Min Range: -0.4739

Max Act: 0.2601. Min Act: -0.0000

Data Index: 1086317 (C4 (Web Text))

Max Activating Token Index: 933