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

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

Neuron 3222 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.4978. Min Act: -0.0002

Data Index: 833076 (C4 (Web Text))

Max Activating Token Index: 741

Text #1

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.6308. Min Act: -0.0002

Data Index: 376562 (C4 (Web Text))

Max Activating Token Index: 115

Text #2

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.6705. Min Act: -0.0002

Data Index: 1282432 (C4 (Web Text))

Max Activating Token Index: 643

Text #3

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.6497. Min Act: -0.0002

Data Index: 939 (C4 (Web Text))

Max Activating Token Index: 771

Text #4

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5340. Min Act: -0.0002

Data Index: 1166353 (C4 (Web Text))

Max Activating Token Index: 846

Text #5

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5351. Min Act: -0.0002

Data Index: 1042089 (C4 (Web Text))

Max Activating Token Index: 944

Text #6

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5389. Min Act: -0.0002

Data Index: 623115 (C4 (Web Text))

Max Activating Token Index: 271

Text #7

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5967. Min Act: -0.0002

Data Index: 279036 (C4 (Web Text))

Max Activating Token Index: 946

Text #8

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5026. Min Act: -0.0002

Data Index: 831004 (C4 (Web Text))

Max Activating Token Index: 889

Text #9

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5079. Min Act: -0.0002

Data Index: 1247296 (C4 (Web Text))

Max Activating Token Index: 589

Text #10

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.7074. Min Act: -0.0002

Data Index: 128749 (C4 (Web Text))

Max Activating Token Index: 594

Text #11

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5304. Min Act: -0.0002

Data Index: 946267 (C4 (Web Text))

Max Activating Token Index: 672

Text #12

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5202. Min Act: -0.0002

Data Index: 1247213 (C4 (Web Text))

Max Activating Token Index: 675

Text #13

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5467. Min Act: -0.0002

Data Index: 1213339 (C4 (Web Text))

Max Activating Token Index: 1012

Text #14

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.4873. Min Act: -0.0002

Data Index: 643687 (C4 (Web Text))

Max Activating Token Index: 125

Text #15

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5018. Min Act: -0.0002

Data Index: 190869 (C4 (Web Text))

Max Activating Token Index: 631

Text #16

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.5475. Min Act: -0.0002

Data Index: 1418664 (Python Code)

Max Activating Token Index: 483

Text #17

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.6507. Min Act: -0.0002

Data Index: 1412964 (Python Code)

Max Activating Token Index: 269

Text #18

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.6307. Min Act: -0.0002

Data Index: 8356 (C4 (Web Text))

Max Activating Token Index: 717

Text #19

Max Range: 0.7074. Min Range: -0.7074

Max Act: 0.4243. Min Act: -0.0002

Data Index: 1208785 (C4 (Web Text))

Max Activating Token Index: 181