Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

Dataset: The Pile

Neuron 57 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l-pile')

Text #0

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.8267. Min Act: -0.0000

Data Index: 1207477 (The Pile)

Max Activating Token Index: 261

Text #1

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7975. Min Act: -0.0000

Data Index: 1080839 (The Pile)

Max Activating Token Index: 94

Text #2

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7655. Min Act: -0.0000

Data Index: 1336052 (The Pile)

Max Activating Token Index: 113

Text #3

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7614. Min Act: -0.0000

Data Index: 214324 (The Pile)

Max Activating Token Index: 202

Text #4

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7778. Min Act: -0.0000

Data Index: 1636298 (The Pile)

Max Activating Token Index: 213

Text #5

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7406. Min Act: -0.0000

Data Index: 1116840 (The Pile)

Max Activating Token Index: 198

Text #6

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7026. Min Act: -0.0000

Data Index: 1593266 (The Pile)

Max Activating Token Index: 222

Text #7

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7244. Min Act: -0.0000

Data Index: 1474357 (The Pile)

Max Activating Token Index: 352

Text #8

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7188. Min Act: -0.0000

Data Index: 340872 (The Pile)

Max Activating Token Index: 413

Text #9

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7347. Min Act: -0.0000

Data Index: 1135413 (The Pile)

Max Activating Token Index: 46

Text #10

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7415. Min Act: -0.0000

Data Index: 1836008 (The Pile)

Max Activating Token Index: 337

Text #11

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.6448. Min Act: -0.0000

Data Index: 659708 (The Pile)

Max Activating Token Index: 224

Text #12

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7032. Min Act: -0.0000

Data Index: 1092722 (The Pile)

Max Activating Token Index: 169

Text #13

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.6710. Min Act: -0.0000

Data Index: 1378340 (The Pile)

Max Activating Token Index: 37

Text #14

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.6977. Min Act: -0.0000

Data Index: 646074 (The Pile)

Max Activating Token Index: 109

Text #15

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7149. Min Act: -0.0000

Data Index: 1166216 (The Pile)

Max Activating Token Index: 169

Text #16

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7105. Min Act: -0.0000

Data Index: 607056 (The Pile)

Max Activating Token Index: 238

Text #17

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7085. Min Act: -0.0000

Data Index: 1831154 (The Pile)

Max Activating Token Index: 672

Text #18

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7281. Min Act: -0.0000

Data Index: 813142 (The Pile)

Max Activating Token Index: 420

Text #19

Max Range: 0.8267. Min Range: -0.8267

Max Act: 0.7078. Min Act: -0.0000

Data Index: 310267 (The Pile)

Max Activating Token Index: 52