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

Dataset: The Pile

Neuron 3754 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.5025. Min Act: -0.0000

Data Index: 759473 (The Pile)

Max Activating Token Index: 666

Text #1

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.6427. Min Act: -0.0000

Data Index: 44686 (The Pile)

Max Activating Token Index: 212

Text #2

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.8022. Min Act: -0.0000

Data Index: 742852 (The Pile)

Max Activating Token Index: 395

Text #3

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.5296. Min Act: -0.0000

Data Index: 203373 (The Pile)

Max Activating Token Index: 330

Text #4

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.5576. Min Act: -0.0000

Data Index: 1884985 (The Pile)

Max Activating Token Index: 401

Text #5

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.5761. Min Act: -0.0000

Data Index: 204253 (The Pile)

Max Activating Token Index: 331

Text #6

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.7135. Min Act: -0.0000

Data Index: 348265 (The Pile)

Max Activating Token Index: 408

Text #7

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.5956. Min Act: -0.0000

Data Index: 1541957 (The Pile)

Max Activating Token Index: 332

Text #8

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.7089. Min Act: -0.0000

Data Index: 613090 (The Pile)

Max Activating Token Index: 1003

Text #9

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.7109. Min Act: -0.0000

Data Index: 881028 (The Pile)

Max Activating Token Index: 439

Text #10

Max Range: 4.0677. Min Range: -4.0677

Max Act: 4.0677. Min Act: -0.0000

Data Index: 1207461 (The Pile)

Max Activating Token Index: 837

Text #11

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.6932. Min Act: -0.0000

Data Index: 1122683 (The Pile)

Max Activating Token Index: 976

Text #12

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.2619. Min Act: -0.0000

Data Index: 747686 (The Pile)

Max Activating Token Index: 993

Text #13

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.2009. Min Act: -0.0000

Data Index: 519793 (The Pile)

Max Activating Token Index: 933

Text #14

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.4837. Min Act: -0.0000

Data Index: 88784 (The Pile)

Max Activating Token Index: 559

Text #15

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.2088. Min Act: -0.0000

Data Index: 1849475 (The Pile)

Max Activating Token Index: 858

Text #16

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.3502. Min Act: -0.0000

Data Index: 754755 (The Pile)

Max Activating Token Index: 603

Text #17

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.1443. Min Act: -0.0000

Data Index: 1194808 (The Pile)

Max Activating Token Index: 415

Text #18

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.4415. Min Act: -0.0000

Data Index: 1429729 (The Pile)

Max Activating Token Index: 825

Text #19

Max Range: 4.0677. Min Range: -4.0677

Max Act: 3.4040. Min Act: -0.0000

Data Index: 165384 (The Pile)

Max Activating Token Index: 838