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

Dataset: The Pile

Neuron 682 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.2699. Min Range: -8.2699

Max Act: 8.2523. Min Act: -0.0000

Data Index: 1796245 (The Pile)

Max Activating Token Index: 452

Text #1

Max Range: 8.2699. Min Range: -8.2699

Max Act: 8.2699. Min Act: -0.0000

Data Index: 1200364 (The Pile)

Max Activating Token Index: 417

Text #2

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.9817. Min Act: -0.0000

Data Index: 983487 (The Pile)

Max Activating Token Index: 486

Text #3

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.7951. Min Act: -0.0000

Data Index: 1611289 (The Pile)

Max Activating Token Index: 680

Text #4

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.6276. Min Act: -0.0000

Data Index: 1162391 (The Pile)

Max Activating Token Index: 466

Text #5

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.5607. Min Act: -0.0000

Data Index: 1187219 (The Pile)

Max Activating Token Index: 631

Text #6

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.7621. Min Act: -0.0000

Data Index: 236955 (The Pile)

Max Activating Token Index: 695

Text #7

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.7364. Min Act: -0.0000

Data Index: 1814000 (The Pile)

Max Activating Token Index: 733

Text #8

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.8938. Min Act: -0.0000

Data Index: 262302 (The Pile)

Max Activating Token Index: 476

Text #9

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.5246. Min Act: -0.0000

Data Index: 1519201 (The Pile)

Max Activating Token Index: 579

Text #10

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.8399. Min Act: -0.0000

Data Index: 1170028 (The Pile)

Max Activating Token Index: 496

Text #11

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.8250. Min Act: -0.0000

Data Index: 1595998 (The Pile)

Max Activating Token Index: 856

Text #12

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.4654. Min Act: -0.0000

Data Index: 90562 (The Pile)

Max Activating Token Index: 643

Text #13

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.5441. Min Act: -0.0000

Data Index: 80936 (The Pile)

Max Activating Token Index: 520

Text #14

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.5454. Min Act: -0.0000

Data Index: 1485080 (The Pile)

Max Activating Token Index: 500

Text #15

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.3651. Min Act: -0.0000

Data Index: 1436482 (The Pile)

Max Activating Token Index: 548

Text #16

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.4751. Min Act: -0.0000

Data Index: 1550974 (The Pile)

Max Activating Token Index: 652

Text #17

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.4827. Min Act: -0.0000

Data Index: 1838792 (The Pile)

Max Activating Token Index: 854

Text #18

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.3355. Min Act: -0.0000

Data Index: 1372467 (The Pile)

Max Activating Token Index: 811

Text #19

Max Range: 8.2699. Min Range: -8.2699

Max Act: 7.6662. Min Act: -0.0000

Data Index: 1488694 (The Pile)

Max Activating Token Index: 219