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

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

Neuron 1546 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.7312. Min Act: -0.0000

Data Index: 1385232 (Python Code)

Max Activating Token Index: 395

Text #1

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.7555. Min Act: -0.0000

Data Index: 1224461 (C4 (Web Text))

Max Activating Token Index: 299

Text #2

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.7446. Min Act: -0.0000

Data Index: 932680 (C4 (Web Text))

Max Activating Token Index: 943

Text #3

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.7172. Min Act: -0.0000

Data Index: 1551407 (Python Code)

Max Activating Token Index: 496

Text #4

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.7027. Min Act: -0.0000

Data Index: 281116 (C4 (Web Text))

Max Activating Token Index: 314

Text #5

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6947. Min Act: -0.0000

Data Index: 11421 (C4 (Web Text))

Max Activating Token Index: 845

Text #6

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6931. Min Act: -0.0000

Data Index: 493640 (C4 (Web Text))

Max Activating Token Index: 980

Text #7

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6630. Min Act: -0.0000

Data Index: 940055 (C4 (Web Text))

Max Activating Token Index: 189

Text #8

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6700. Min Act: -0.0000

Data Index: 1605960 (Python Code)

Max Activating Token Index: 547

Text #9

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6120. Min Act: -0.0000

Data Index: 560060 (C4 (Web Text))

Max Activating Token Index: 982

Text #10

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6039. Min Act: -0.0000

Data Index: 1056038 (C4 (Web Text))

Max Activating Token Index: 862

Text #11

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6288. Min Act: -0.0000

Data Index: 895536 (C4 (Web Text))

Max Activating Token Index: 345

Text #12

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6454. Min Act: -0.0000

Data Index: 195737 (C4 (Web Text))

Max Activating Token Index: 1018

Text #13

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6198. Min Act: -0.0000

Data Index: 1059435 (C4 (Web Text))

Max Activating Token Index: 425

Text #14

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.5881. Min Act: 0.0001

Data Index: 1540564 (Python Code)

Max Activating Token Index: 210

Text #15

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6259. Min Act: -0.0000

Data Index: 1222943 (C4 (Web Text))

Max Activating Token Index: 322

Text #16

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6049. Min Act: -0.0000

Data Index: 1363309 (Python Code)

Max Activating Token Index: 483

Text #17

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6224. Min Act: -0.0000

Data Index: 1633326 (Python Code)

Max Activating Token Index: 457

Text #18

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6820. Min Act: -0.0000

Data Index: 1364399 (Python Code)

Max Activating Token Index: 990

Text #19

Max Range: 0.7555. Min Range: -0.7555

Max Act: 0.6091. Min Act: -0.0000

Data Index: 526530 (C4 (Web Text))

Max Activating Token Index: 909