Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 65 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.4052. Min Range: -1.4052

Max Act: 1.4052. Min Act: -0.0002

Data Index: 67453 (C4 (Web Text))

Max Activating Token Index: 342

Text #1

Max Range: 1.4052. Min Range: -1.4052

Max Act: 1.2247. Min Act: -0.0002

Data Index: 455476 (C4 (Web Text))

Max Activating Token Index: 898

Text #2

Max Range: 1.4052. Min Range: -1.4052

Max Act: 1.0031. Min Act: -0.0002

Data Index: 1315481 (C4 (Web Text))

Max Activating Token Index: 986

Text #3

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.9457. Min Act: -0.0002

Data Index: 1215754 (C4 (Web Text))

Max Activating Token Index: 241

Text #4

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.9687. Min Act: -0.0002

Data Index: 347452 (C4 (Web Text))

Max Activating Token Index: 476

Text #5

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.9636. Min Act: -0.0002

Data Index: 514693 (C4 (Web Text))

Max Activating Token Index: 169

Text #6

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.9184. Min Act: -0.0002

Data Index: 1265105 (C4 (Web Text))

Max Activating Token Index: 779

Text #7

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.9114. Min Act: -0.0002

Data Index: 392742 (C4 (Web Text))

Max Activating Token Index: 769

Text #8

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8954. Min Act: -0.0002

Data Index: 22158 (C4 (Web Text))

Max Activating Token Index: 10

Text #9

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8812. Min Act: -0.0002

Data Index: 1047202 (C4 (Web Text))

Max Activating Token Index: 441

Text #10

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8324. Min Act: -0.0002

Data Index: 307080 (C4 (Web Text))

Max Activating Token Index: 230

Text #11

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8292. Min Act: -0.0002

Data Index: 1011641 (C4 (Web Text))

Max Activating Token Index: 436

Text #12

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8279. Min Act: -0.0002

Data Index: 1575819 (Python Code)

Max Activating Token Index: 791

Text #13

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8683. Min Act: -0.0002

Data Index: 436864 (C4 (Web Text))

Max Activating Token Index: 839

Text #14

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.7921. Min Act: -0.0002

Data Index: 1308274 (C4 (Web Text))

Max Activating Token Index: 714

Text #15

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8499. Min Act: -0.0002

Data Index: 1149017 (C4 (Web Text))

Max Activating Token Index: 723

Text #16

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8517. Min Act: -0.0002

Data Index: 351978 (C4 (Web Text))

Max Activating Token Index: 89

Text #17

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8869. Min Act: -0.0002

Data Index: 447366 (C4 (Web Text))

Max Activating Token Index: 292

Text #18

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8069. Min Act: -0.0002

Data Index: 519332 (C4 (Web Text))

Max Activating Token Index: 68

Text #19

Max Range: 1.4052. Min Range: -1.4052

Max Act: 0.8044. Min Act: -0.0002

Data Index: 1168452 (C4 (Web Text))

Max Activating Token Index: 82