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

Dataset: The Pile

Neuron 929 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: 4.4683. Min Range: -4.4683

Max Act: 4.4683. Min Act: -0.0000

Data Index: 567665 (The Pile)

Max Activating Token Index: 626

Text #1

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.9876. Min Act: -0.0000

Data Index: 1975828 (The Pile)

Max Activating Token Index: 638

Text #2

Max Range: 4.4683. Min Range: -4.4683

Max Act: 4.2314. Min Act: -0.0000

Data Index: 192889 (The Pile)

Max Activating Token Index: 546

Text #3

Max Range: 4.4683. Min Range: -4.4683

Max Act: 4.0018. Min Act: -0.0000

Data Index: 740138 (The Pile)

Max Activating Token Index: 683

Text #4

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.9307. Min Act: -0.0000

Data Index: 1860905 (The Pile)

Max Activating Token Index: 301

Text #5

Max Range: 4.4683. Min Range: -4.4683

Max Act: 4.0414. Min Act: -0.0000

Data Index: 1790477 (The Pile)

Max Activating Token Index: 807

Text #6

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.9020. Min Act: -0.0000

Data Index: 1637287 (The Pile)

Max Activating Token Index: 851

Text #7

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.8574. Min Act: -0.0000

Data Index: 1663365 (The Pile)

Max Activating Token Index: 486

Text #8

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.7236. Min Act: -0.0000

Data Index: 1735827 (The Pile)

Max Activating Token Index: 575

Text #9

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.5467. Min Act: -0.0000

Data Index: 1947474 (The Pile)

Max Activating Token Index: 548

Text #10

Max Range: 4.4683. Min Range: -4.4683

Max Act: 4.0122. Min Act: -0.0000

Data Index: 1611744 (The Pile)

Max Activating Token Index: 529

Text #11

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.7082. Min Act: -0.0000

Data Index: 418111 (The Pile)

Max Activating Token Index: 540

Text #12

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.5219. Min Act: -0.0000

Data Index: 1866922 (The Pile)

Max Activating Token Index: 653

Text #13

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.7841. Min Act: -0.0000

Data Index: 1016610 (The Pile)

Max Activating Token Index: 303

Text #14

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.5919. Min Act: -0.0000

Data Index: 1411381 (The Pile)

Max Activating Token Index: 538

Text #15

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.8836. Min Act: -0.0000

Data Index: 1041195 (The Pile)

Max Activating Token Index: 389

Text #16

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.8706. Min Act: -0.0000

Data Index: 353577 (The Pile)

Max Activating Token Index: 637

Text #17

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.5326. Min Act: -0.0000

Data Index: 1475978 (The Pile)

Max Activating Token Index: 627

Text #18

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.5433. Min Act: -0.0000

Data Index: 1691207 (The Pile)

Max Activating Token Index: 548

Text #19

Max Range: 4.4683. Min Range: -4.4683

Max Act: 3.5280. Min Act: -0.0000

Data Index: 51969 (The Pile)

Max Activating Token Index: 973