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

Dataset: The Pile

Neuron 5084 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 10.2065. Min Range: -10.2065

Max Act: 10.1753. Min Act: -0.0000

Data Index: 201760 (The Pile)

Max Activating Token Index: 39

Text #1

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.8637. Min Act: -0.0000

Data Index: 1313566 (The Pile)

Max Activating Token Index: 217

Text #2

Max Range: 10.2065. Min Range: -10.2065

Max Act: 10.1199. Min Act: -0.0000

Data Index: 86918 (The Pile)

Max Activating Token Index: 379

Text #3

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.9846. Min Act: -0.0000

Data Index: 214657 (The Pile)

Max Activating Token Index: 100

Text #4

Max Range: 10.2065. Min Range: -10.2065

Max Act: 10.2065. Min Act: -0.0000

Data Index: 1491444 (The Pile)

Max Activating Token Index: 83

Text #5

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.8241. Min Act: -0.0000

Data Index: 695335 (The Pile)

Max Activating Token Index: 546

Text #6

Max Range: 10.2065. Min Range: -10.2065

Max Act: 10.1448. Min Act: -0.0000

Data Index: 1512992 (The Pile)

Max Activating Token Index: 115

Text #7

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.8492. Min Act: -0.0000

Data Index: 1367036 (The Pile)

Max Activating Token Index: 221

Text #8

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.9787. Min Act: -0.0000

Data Index: 1278168 (The Pile)

Max Activating Token Index: 149

Text #9

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.8129. Min Act: -0.0000

Data Index: 1474290 (The Pile)

Max Activating Token Index: 79

Text #10

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.8176. Min Act: -0.0000

Data Index: 984963 (The Pile)

Max Activating Token Index: 808

Text #11

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.7963. Min Act: -0.0000

Data Index: 1834063 (The Pile)

Max Activating Token Index: 625

Text #12

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.8059. Min Act: -0.0000

Data Index: 1970760 (The Pile)

Max Activating Token Index: 103

Text #13

Max Range: 10.2065. Min Range: -10.2065

Max Act: 10.0490. Min Act: -0.0000

Data Index: 715619 (The Pile)

Max Activating Token Index: 75

Text #14

Max Range: 10.2065. Min Range: -10.2065

Max Act: 10.0406. Min Act: -0.0000

Data Index: 1003734 (The Pile)

Max Activating Token Index: 184

Text #15

Max Range: 10.2065. Min Range: -10.2065

Max Act: 10.0695. Min Act: -0.0000

Data Index: 201784 (The Pile)

Max Activating Token Index: 672

Text #16

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.8934. Min Act: -0.0000

Data Index: 212097 (The Pile)

Max Activating Token Index: 202

Text #17

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.8818. Min Act: -0.0000

Data Index: 1048646 (The Pile)

Max Activating Token Index: 609

Text #18

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.7657. Min Act: -0.0000

Data Index: 646235 (The Pile)

Max Activating Token Index: 267

Text #19

Max Range: 10.2065. Min Range: -10.2065

Max Act: 9.8807. Min Act: -0.0000

Data Index: 167973 (The Pile)

Max Activating Token Index: 596