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

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 4259 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.4820. Min Act: -0.0000

Data Index: 1278106 (C4 (Web Text))

Max Activating Token Index: 923

Text #1

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.4643. Min Act: -0.0000

Data Index: 1386334 (Python Code)

Max Activating Token Index: 407

Text #2

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.4641. Min Act: -0.0000

Data Index: 1585449 (Python Code)

Max Activating Token Index: 777

Text #3

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.4182. Min Act: -0.0000

Data Index: 803400 (C4 (Web Text))

Max Activating Token Index: 986

Text #4

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.4127. Min Act: -0.0000

Data Index: 1616620 (Python Code)

Max Activating Token Index: 717

Text #5

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3808. Min Act: -0.0000

Data Index: 1611931 (Python Code)

Max Activating Token Index: 686

Text #6

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.4178. Min Act: -0.0000

Data Index: 577833 (C4 (Web Text))

Max Activating Token Index: 787

Text #7

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3896. Min Act: -0.0000

Data Index: 1505614 (Python Code)

Max Activating Token Index: 667

Text #8

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3712. Min Act: -0.0000

Data Index: 1376951 (Python Code)

Max Activating Token Index: 355

Text #9

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3605. Min Act: -0.0000

Data Index: 1621994 (Python Code)

Max Activating Token Index: 190

Text #10

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3868. Min Act: -0.0000

Data Index: 1144749 (C4 (Web Text))

Max Activating Token Index: 942

Text #11

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3841. Min Act: -0.0000

Data Index: 923568 (C4 (Web Text))

Max Activating Token Index: 457

Text #12

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3821. Min Act: -0.0000

Data Index: 1507890 (Python Code)

Max Activating Token Index: 969

Text #13

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3676. Min Act: -0.0000

Data Index: 84615 (C4 (Web Text))

Max Activating Token Index: 849

Text #14

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3518. Min Act: -0.0000

Data Index: 1378278 (Python Code)

Max Activating Token Index: 683

Text #15

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3481. Min Act: -0.0000

Data Index: 885508 (C4 (Web Text))

Max Activating Token Index: 814

Text #16

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3809. Min Act: -0.0000

Data Index: 415343 (C4 (Web Text))

Max Activating Token Index: 606

Text #17

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3477. Min Act: -0.0000

Data Index: 1418922 (Python Code)

Max Activating Token Index: 363

Text #18

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3535. Min Act: -0.0000

Data Index: 1629378 (Python Code)

Max Activating Token Index: 336

Text #19

Max Range: 0.4820. Min Range: -0.4820

Max Act: 0.3689. Min Act: -0.0000

Data Index: 1530802 (Python Code)

Max Activating Token Index: 699