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

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

Neuron 3117 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.4229. Min Range: -4.4229

Max Act: 4.4229. Min Act: -0.0001

Data Index: 702901 (C4 (Web Text))

Max Activating Token Index: 233

Text #1

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.7049. Min Act: -0.0001

Data Index: 1480816 (Python Code)

Max Activating Token Index: 216

Text #2

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.4524. Min Act: -0.0001

Data Index: 695329 (C4 (Web Text))

Max Activating Token Index: 768

Text #3

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.5467. Min Act: -0.0001

Data Index: 579827 (C4 (Web Text))

Max Activating Token Index: 668

Text #4

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.3806. Min Act: -0.0001

Data Index: 1536229 (Python Code)

Max Activating Token Index: 34

Text #5

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.1814. Min Act: -0.0001

Data Index: 1547255 (Python Code)

Max Activating Token Index: 509

Text #6

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.3935. Min Act: -0.0001

Data Index: 1146969 (C4 (Web Text))

Max Activating Token Index: 278

Text #7

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.3781. Min Act: -0.0001

Data Index: 1465055 (Python Code)

Max Activating Token Index: 6

Text #8

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.4014. Min Act: -0.0001

Data Index: 435820 (C4 (Web Text))

Max Activating Token Index: 723

Text #9

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.2098. Min Act: -0.0001

Data Index: 114411 (C4 (Web Text))

Max Activating Token Index: 528

Text #10

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.1849. Min Act: -0.0001

Data Index: 807059 (C4 (Web Text))

Max Activating Token Index: 930

Text #11

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.4483. Min Act: -0.0001

Data Index: 583280 (C4 (Web Text))

Max Activating Token Index: 903

Text #12

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.3162. Min Act: -0.0001

Data Index: 959686 (C4 (Web Text))

Max Activating Token Index: 920

Text #13

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.2396. Min Act: -0.0001

Data Index: 1179855 (C4 (Web Text))

Max Activating Token Index: 368

Text #14

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.0885. Min Act: -0.0001

Data Index: 290367 (C4 (Web Text))

Max Activating Token Index: 7

Text #15

Max Range: 4.4229. Min Range: -4.4229

Max Act: 3.0673. Min Act: -0.0001

Data Index: 715881 (C4 (Web Text))

Max Activating Token Index: 583

Text #16

Max Range: 4.4229. Min Range: -4.4229

Max Act: 2.7688. Min Act: -0.0001

Data Index: 1485525 (Python Code)

Max Activating Token Index: 840

Text #17

Max Range: 4.4229. Min Range: -4.4229

Max Act: 2.9763. Min Act: -0.0001

Data Index: 147081 (C4 (Web Text))

Max Activating Token Index: 118

Text #18

Max Range: 4.4229. Min Range: -4.4229

Max Act: 2.9707. Min Act: -0.0001

Data Index: 405257 (C4 (Web Text))

Max Activating Token Index: 509

Text #19

Max Range: 4.4229. Min Range: -4.4229

Max Act: 2.9637. Min Act: -0.0001

Data Index: 194510 (C4 (Web Text))

Max Activating Token Index: 325