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

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

Neuron 2654 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.5030. Min Act: -0.0001

Data Index: 278619 (C4 (Web Text))

Max Activating Token Index: 720

Text #1

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.4513. Min Act: -0.0001

Data Index: 28626 (C4 (Web Text))

Max Activating Token Index: 524

Text #2

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3731. Min Act: -0.0001

Data Index: 475930 (C4 (Web Text))

Max Activating Token Index: 247

Text #3

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3592. Min Act: -0.0000

Data Index: 1126929 (C4 (Web Text))

Max Activating Token Index: 816

Text #4

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3708. Min Act: -0.0001

Data Index: 1027432 (C4 (Web Text))

Max Activating Token Index: 910

Text #5

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3967. Min Act: -0.0001

Data Index: 1083153 (C4 (Web Text))

Max Activating Token Index: 397

Text #6

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3545. Min Act: -0.0001

Data Index: 1596838 (Python Code)

Max Activating Token Index: 252

Text #7

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3954. Min Act: -0.0001

Data Index: 779105 (C4 (Web Text))

Max Activating Token Index: 721

Text #8

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3474. Min Act: -0.0001

Data Index: 635580 (C4 (Web Text))

Max Activating Token Index: 167

Text #9

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3125. Min Act: -0.0001

Data Index: 1168884 (C4 (Web Text))

Max Activating Token Index: 298

Text #10

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3090. Min Act: -0.0001

Data Index: 1365011 (Python Code)

Max Activating Token Index: 642

Text #11

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3145. Min Act: -0.0000

Data Index: 152215 (C4 (Web Text))

Max Activating Token Index: 430

Text #12

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.2982. Min Act: -0.0001

Data Index: 545196 (C4 (Web Text))

Max Activating Token Index: 928

Text #13

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.3097. Min Act: -0.0000

Data Index: 539671 (C4 (Web Text))

Max Activating Token Index: 562

Text #14

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.2974. Min Act: -0.0001

Data Index: 1344116 (C4 (Web Text))

Max Activating Token Index: 269

Text #15

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.2935. Min Act: -0.0000

Data Index: 1407959 (Python Code)

Max Activating Token Index: 520

Text #16

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.2865. Min Act: -0.0001

Data Index: 619456 (C4 (Web Text))

Max Activating Token Index: 821

Text #17

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.2800. Min Act: -0.0001

Data Index: 1447599 (Python Code)

Max Activating Token Index: 919

Text #18

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.2833. Min Act: -0.0001

Data Index: 1425983 (Python Code)

Max Activating Token Index: 386

Text #19

Max Range: 0.5030. Min Range: -0.5030

Max Act: 0.2634. Min Act: -0.0001

Data Index: 1605712 (Python Code)

Max Activating Token Index: 718