Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2149 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.7201. Min Act: -0.0002

Data Index: 638127 (C4 (Web Text))

Max Activating Token Index: 984

Text #1

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.7139. Min Act: -0.0001

Data Index: 406942 (C4 (Web Text))

Max Activating Token Index: 229

Text #2

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.7200. Min Act: -0.0002

Data Index: 1163682 (C4 (Web Text))

Max Activating Token Index: 651

Text #3

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6365. Min Act: -0.0002

Data Index: 295752 (C4 (Web Text))

Max Activating Token Index: 486

Text #4

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6775. Min Act: -0.0001

Data Index: 1196505 (C4 (Web Text))

Max Activating Token Index: 306

Text #5

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6508. Min Act: -0.0002

Data Index: 932365 (C4 (Web Text))

Max Activating Token Index: 577

Text #6

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6299. Min Act: -0.0002

Data Index: 86186 (C4 (Web Text))

Max Activating Token Index: 617

Text #7

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.5514. Min Act: -0.0001

Data Index: 1563261 (Python Code)

Max Activating Token Index: 678

Text #8

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6820. Min Act: -0.0001

Data Index: 1561850 (Python Code)

Max Activating Token Index: 14

Text #9

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6063. Min Act: -0.0001

Data Index: 1496622 (Python Code)

Max Activating Token Index: 490

Text #10

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6409. Min Act: -0.0002

Data Index: 861982 (C4 (Web Text))

Max Activating Token Index: 234

Text #11

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6569. Min Act: -0.0002

Data Index: 731429 (C4 (Web Text))

Max Activating Token Index: 1009

Text #12

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6174. Min Act: -0.0001

Data Index: 233068 (C4 (Web Text))

Max Activating Token Index: 12

Text #13

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6097. Min Act: -0.0002

Data Index: 1292087 (C4 (Web Text))

Max Activating Token Index: 300

Text #14

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6259. Min Act: -0.0001

Data Index: 380282 (C4 (Web Text))

Max Activating Token Index: 59

Text #15

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6279. Min Act: -0.0001

Data Index: 665844 (C4 (Web Text))

Max Activating Token Index: 640

Text #16

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6367. Min Act: -0.0002

Data Index: 983332 (C4 (Web Text))

Max Activating Token Index: 965

Text #17

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.5826. Min Act: -0.0002

Data Index: 989119 (C4 (Web Text))

Max Activating Token Index: 307

Text #18

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.6204. Min Act: -0.0001

Data Index: 1650211 (Python Code)

Max Activating Token Index: 943

Text #19

Max Range: 0.7201. Min Range: -0.7201

Max Act: 0.5806. Min Act: -0.0002

Data Index: 993899 (C4 (Web Text))

Max Activating Token Index: 536