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

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

Neuron 3763 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: 0.6970. Min Range: -0.6970

Max Act: 0.6970. Min Act: -0.0000

Data Index: 217677 (C4 (Web Text))

Max Activating Token Index: 708

Text #1

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.6614. Min Act: -0.0000

Data Index: 561094 (C4 (Web Text))

Max Activating Token Index: 292

Text #2

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.6711. Min Act: -0.0000

Data Index: 608767 (C4 (Web Text))

Max Activating Token Index: 166

Text #3

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.6510. Min Act: -0.0000

Data Index: 6152 (C4 (Web Text))

Max Activating Token Index: 138

Text #4

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5996. Min Act: -0.0000

Data Index: 245491 (C4 (Web Text))

Max Activating Token Index: 598

Text #5

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.6206. Min Act: -0.0000

Data Index: 719519 (C4 (Web Text))

Max Activating Token Index: 109

Text #6

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.6009. Min Act: -0.0000

Data Index: 1206296 (C4 (Web Text))

Max Activating Token Index: 69

Text #7

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.6096. Min Act: -0.0000

Data Index: 526063 (C4 (Web Text))

Max Activating Token Index: 326

Text #8

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5483. Min Act: -0.0000

Data Index: 1080810 (C4 (Web Text))

Max Activating Token Index: 408

Text #9

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5334. Min Act: -0.0000

Data Index: 1351581 (C4 (Web Text))

Max Activating Token Index: 200

Text #10

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5444. Min Act: -0.0000

Data Index: 948534 (C4 (Web Text))

Max Activating Token Index: 325

Text #11

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5426. Min Act: -0.0000

Data Index: 551875 (C4 (Web Text))

Max Activating Token Index: 540

Text #12

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5409. Min Act: -0.0000

Data Index: 86904 (C4 (Web Text))

Max Activating Token Index: 342

Text #13

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5557. Min Act: -0.0000

Data Index: 875344 (C4 (Web Text))

Max Activating Token Index: 441

Text #14

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5261. Min Act: -0.0000

Data Index: 936481 (C4 (Web Text))

Max Activating Token Index: 272

Text #15

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.4923. Min Act: -0.0000

Data Index: 1355874 (C4 (Web Text))

Max Activating Token Index: 652

Text #16

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5253. Min Act: -0.0000

Data Index: 981187 (C4 (Web Text))

Max Activating Token Index: 688

Text #17

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.5045. Min Act: -0.0000

Data Index: 309895 (C4 (Web Text))

Max Activating Token Index: 430

Text #18

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.4845. Min Act: -0.0000

Data Index: 803592 (C4 (Web Text))

Max Activating Token Index: 1013

Text #19

Max Range: 0.6970. Min Range: -0.6970

Max Act: 0.4780. Min Act: -0.0000

Data Index: 1192105 (C4 (Web Text))

Max Activating Token Index: 398