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

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

Neuron 3699 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.9853. Min Act: -0.0001

Data Index: 1078074 (C4 (Web Text))

Max Activating Token Index: 60

Text #1

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.7647. Min Act: -0.0001

Data Index: 77207 (C4 (Web Text))

Max Activating Token Index: 190

Text #2

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.6996. Min Act: -0.0001

Data Index: 458543 (C4 (Web Text))

Max Activating Token Index: 137

Text #3

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.8737. Min Act: -0.0001

Data Index: 397767 (C4 (Web Text))

Max Activating Token Index: 116

Text #4

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.5929. Min Act: -0.0001

Data Index: 1644060 (Python Code)

Max Activating Token Index: 355

Text #5

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.7582. Min Act: -0.0001

Data Index: 964593 (C4 (Web Text))

Max Activating Token Index: 316

Text #6

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.8582. Min Act: -0.0001

Data Index: 519389 (C4 (Web Text))

Max Activating Token Index: 1001

Text #7

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.4497. Min Act: -0.0001

Data Index: 198859 (C4 (Web Text))

Max Activating Token Index: 216

Text #8

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.6917. Min Act: -0.0001

Data Index: 1344518 (C4 (Web Text))

Max Activating Token Index: 185

Text #9

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.8725. Min Act: -0.0001

Data Index: 836011 (C4 (Web Text))

Max Activating Token Index: 11

Text #10

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.6930. Min Act: -0.0001

Data Index: 1370160 (Python Code)

Max Activating Token Index: 702

Text #11

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.8087. Min Act: -0.0001

Data Index: 210940 (C4 (Web Text))

Max Activating Token Index: 139

Text #12

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.6767. Min Act: -0.0001

Data Index: 657381 (C4 (Web Text))

Max Activating Token Index: 381

Text #13

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.6888. Min Act: -0.0001

Data Index: 708091 (C4 (Web Text))

Max Activating Token Index: 142

Text #14

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.7365. Min Act: -0.0001

Data Index: 1318499 (C4 (Web Text))

Max Activating Token Index: 330

Text #15

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.7561. Min Act: -0.0002

Data Index: 1148694 (C4 (Web Text))

Max Activating Token Index: 929

Text #16

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.7530. Min Act: -0.0001

Data Index: 222416 (C4 (Web Text))

Max Activating Token Index: 221

Text #17

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.7510. Min Act: -0.0001

Data Index: 15434 (C4 (Web Text))

Max Activating Token Index: 58

Text #18

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.6295. Min Act: -0.0001

Data Index: 1466527 (Python Code)

Max Activating Token Index: 323

Text #19

Max Range: 0.9853. Min Range: -0.9853

Max Act: 0.6354. Min Act: -0.0001

Data Index: 434615 (C4 (Web Text))

Max Activating Token Index: 668