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

Dataset: The Pile

Neuron 3575 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.5632. Min Range: -7.5632

Max Act: 7.3396. Min Act: -0.0000

Data Index: 1287387 (The Pile)

Max Activating Token Index: 257

Text #1

Max Range: 7.5632. Min Range: -7.5632

Max Act: 7.3198. Min Act: -0.0000

Data Index: 1633796 (The Pile)

Max Activating Token Index: 374

Text #2

Max Range: 7.5632. Min Range: -7.5632

Max Act: 7.5632. Min Act: -0.0000

Data Index: 1220841 (The Pile)

Max Activating Token Index: 685

Text #3

Max Range: 7.5632. Min Range: -7.5632

Max Act: 7.0366. Min Act: -0.0000

Data Index: 1477473 (The Pile)

Max Activating Token Index: 307

Text #4

Max Range: 7.5632. Min Range: -7.5632

Max Act: 7.2740. Min Act: -0.0000

Data Index: 1751017 (The Pile)

Max Activating Token Index: 541

Text #5

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.9846. Min Act: -0.0000

Data Index: 881781 (The Pile)

Max Activating Token Index: 382

Text #6

Max Range: 7.5632. Min Range: -7.5632

Max Act: 7.1371. Min Act: -0.0000

Data Index: 266252 (The Pile)

Max Activating Token Index: 383

Text #7

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.8481. Min Act: -0.0000

Data Index: 1713078 (The Pile)

Max Activating Token Index: 387

Text #8

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.4249. Min Act: -0.0000

Data Index: 1539361 (The Pile)

Max Activating Token Index: 515

Text #9

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.7113. Min Act: -0.0000

Data Index: 1728447 (The Pile)

Max Activating Token Index: 555

Text #10

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.1429. Min Act: -0.0000

Data Index: 881223 (The Pile)

Max Activating Token Index: 983

Text #11

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.2923. Min Act: -0.0000

Data Index: 659858 (The Pile)

Max Activating Token Index: 729

Text #12

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.4571. Min Act: -0.0000

Data Index: 616358 (The Pile)

Max Activating Token Index: 1009

Text #13

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.2124. Min Act: -0.0000

Data Index: 1485716 (The Pile)

Max Activating Token Index: 586

Text #14

Max Range: 7.5632. Min Range: -7.5632

Max Act: 5.7634. Min Act: -0.0000

Data Index: 828645 (The Pile)

Max Activating Token Index: 450

Text #15

Max Range: 7.5632. Min Range: -7.5632

Max Act: 5.7004. Min Act: -0.0000

Data Index: 198370 (The Pile)

Max Activating Token Index: 595

Text #16

Max Range: 7.5632. Min Range: -7.5632

Max Act: 5.8615. Min Act: -0.0000

Data Index: 258614 (The Pile)

Max Activating Token Index: 522

Text #17

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.0294. Min Act: -0.0000

Data Index: 501381 (The Pile)

Max Activating Token Index: 554

Text #18

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.4176. Min Act: -0.0000

Data Index: 185925 (The Pile)

Max Activating Token Index: 196

Text #19

Max Range: 7.5632. Min Range: -7.5632

Max Act: 6.0513. Min Act: -0.0000

Data Index: 316334 (The Pile)

Max Activating Token Index: 463