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

Dataset: The Pile

Neuron 1096 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: 5.3333. Min Range: -5.3333

Max Act: 5.3333. Min Act: -0.0000

Data Index: 1722241 (The Pile)

Max Activating Token Index: 569

Text #1

Max Range: 5.3333. Min Range: -5.3333

Max Act: 5.0299. Min Act: -0.0000

Data Index: 1922412 (The Pile)

Max Activating Token Index: 72

Text #2

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.5893. Min Act: -0.0000

Data Index: 1785531 (The Pile)

Max Activating Token Index: 333

Text #3

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.7373. Min Act: -0.0000

Data Index: 1252055 (The Pile)

Max Activating Token Index: 197

Text #4

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.6448. Min Act: -0.0000

Data Index: 719933 (The Pile)

Max Activating Token Index: 24

Text #5

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.3117. Min Act: -0.0000

Data Index: 1968018 (The Pile)

Max Activating Token Index: 544

Text #6

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.4381. Min Act: -0.0000

Data Index: 1859602 (The Pile)

Max Activating Token Index: 165

Text #7

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.3577. Min Act: -0.0000

Data Index: 618449 (The Pile)

Max Activating Token Index: 389

Text #8

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.4453. Min Act: -0.0000

Data Index: 387659 (The Pile)

Max Activating Token Index: 495

Text #9

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.2228. Min Act: -0.0000

Data Index: 1956810 (The Pile)

Max Activating Token Index: 441

Text #10

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.2031. Min Act: -0.0000

Data Index: 9265 (The Pile)

Max Activating Token Index: 178

Text #11

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.1656. Min Act: -0.0000

Data Index: 502236 (The Pile)

Max Activating Token Index: 237

Text #12

Max Range: 5.3333. Min Range: -5.3333

Max Act: 3.7742. Min Act: -0.0000

Data Index: 1714079 (The Pile)

Max Activating Token Index: 844

Text #13

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.2247. Min Act: -0.0000

Data Index: 224329 (The Pile)

Max Activating Token Index: 702

Text #14

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.0974. Min Act: -0.0000

Data Index: 756452 (The Pile)

Max Activating Token Index: 760

Text #15

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.0840. Min Act: -0.0000

Data Index: 1658380 (The Pile)

Max Activating Token Index: 200

Text #16

Max Range: 5.3333. Min Range: -5.3333

Max Act: 4.0353. Min Act: -0.0000

Data Index: 1797636 (The Pile)

Max Activating Token Index: 266

Text #17

Max Range: 5.3333. Min Range: -5.3333

Max Act: 3.4960. Min Act: -0.0000

Data Index: 1015229 (The Pile)

Max Activating Token Index: 432

Text #18

Max Range: 5.3333. Min Range: -5.3333

Max Act: 3.7840. Min Act: -0.0000

Data Index: 729106 (The Pile)

Max Activating Token Index: 509

Text #19

Max Range: 5.3333. Min Range: -5.3333

Max Act: 3.7967. Min Act: 0.0000

Data Index: 1997761 (The Pile)

Max Activating Token Index: 486