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

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

Neuron 2636 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: 2.6145. Min Range: -2.6145

Max Act: 2.6145. Min Act: -0.0000

Data Index: 363387 (C4 (Web Text))

Max Activating Token Index: 415

Text #1

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.9677. Min Act: -0.0000

Data Index: 1222599 (C4 (Web Text))

Max Activating Token Index: 513

Text #2

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.6407. Min Act: -0.0000

Data Index: 884261 (C4 (Web Text))

Max Activating Token Index: 972

Text #3

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.5705. Min Act: -0.0000

Data Index: 588351 (C4 (Web Text))

Max Activating Token Index: 176

Text #4

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.5550. Min Act: -0.0000

Data Index: 1033520 (C4 (Web Text))

Max Activating Token Index: 228

Text #5

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.5179. Min Act: -0.0000

Data Index: 1039368 (C4 (Web Text))

Max Activating Token Index: 625

Text #6

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.5239. Min Act: -0.0000

Data Index: 552548 (C4 (Web Text))

Max Activating Token Index: 502

Text #7

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.5049. Min Act: -0.0000

Data Index: 436520 (C4 (Web Text))

Max Activating Token Index: 54

Text #8

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.5830. Min Act: -0.0000

Data Index: 1492180 (Python Code)

Max Activating Token Index: 625

Text #9

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.4924. Min Act: -0.0000

Data Index: 1092174 (C4 (Web Text))

Max Activating Token Index: 667

Text #10

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.4594. Min Act: -0.0000

Data Index: 1200449 (C4 (Web Text))

Max Activating Token Index: 947

Text #11

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.4434. Min Act: -0.0000

Data Index: 1485790 (Python Code)

Max Activating Token Index: 414

Text #12

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.4155. Min Act: -0.0000

Data Index: 449770 (C4 (Web Text))

Max Activating Token Index: 98

Text #13

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.4051. Min Act: -0.0000

Data Index: 769558 (C4 (Web Text))

Max Activating Token Index: 993

Text #14

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.3029. Min Act: -0.0000

Data Index: 175098 (C4 (Web Text))

Max Activating Token Index: 872

Text #15

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.3581. Min Act: -0.0000

Data Index: 270361 (C4 (Web Text))

Max Activating Token Index: 240

Text #16

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.2819. Min Act: -0.0000

Data Index: 447030 (C4 (Web Text))

Max Activating Token Index: 458

Text #17

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.3642. Min Act: -0.0000

Data Index: 1125142 (C4 (Web Text))

Max Activating Token Index: 135

Text #18

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.3525. Min Act: -0.0000

Data Index: 1047918 (C4 (Web Text))

Max Activating Token Index: 504

Text #19

Max Range: 2.6145. Min Range: -2.6145

Max Act: 1.3303. Min Act: -0.0000

Data Index: 306802 (C4 (Web Text))

Max Activating Token Index: 111