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

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

Neuron 3548 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7430. Min Range: -3.7430

Max Act: 3.7430. Min Act: -0.0001

Data Index: 1607809 (Python Code)

Max Activating Token Index: 25

Text #1

Max Range: 3.7430. Min Range: -3.7430

Max Act: 3.4524. Min Act: -0.0001

Data Index: 255338 (C4 (Web Text))

Max Activating Token Index: 269

Text #2

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.9119. Min Act: -0.0001

Data Index: 153700 (C4 (Web Text))

Max Activating Token Index: 102

Text #3

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.8947. Min Act: -0.0001

Data Index: 794154 (C4 (Web Text))

Max Activating Token Index: 564

Text #4

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.7090. Min Act: -0.0000

Data Index: 1415655 (Python Code)

Max Activating Token Index: 179

Text #5

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.5360. Min Act: -0.0001

Data Index: 935435 (C4 (Web Text))

Max Activating Token Index: 964

Text #6

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.7435. Min Act: -0.0001

Data Index: 270070 (C4 (Web Text))

Max Activating Token Index: 416

Text #7

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.7251. Min Act: -0.0001

Data Index: 1353105 (C4 (Web Text))

Max Activating Token Index: 796

Text #8

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.7206. Min Act: -0.0001

Data Index: 1507988 (Python Code)

Max Activating Token Index: 141

Text #9

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.7091. Min Act: -0.0001

Data Index: 702046 (C4 (Web Text))

Max Activating Token Index: 133

Text #10

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.6974. Min Act: -0.0001

Data Index: 210571 (C4 (Web Text))

Max Activating Token Index: 277

Text #11

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.8196. Min Act: -0.0001

Data Index: 187981 (C4 (Web Text))

Max Activating Token Index: 317

Text #12

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.5593. Min Act: -0.0001

Data Index: 1048760 (C4 (Web Text))

Max Activating Token Index: 299

Text #13

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.5123. Min Act: -0.0001

Data Index: 492484 (C4 (Web Text))

Max Activating Token Index: 873

Text #14

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.4899. Min Act: -0.0000

Data Index: 1633302 (Python Code)

Max Activating Token Index: 315

Text #15

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.5941. Min Act: -0.0001

Data Index: 19751 (C4 (Web Text))

Max Activating Token Index: 414

Text #16

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.6810. Min Act: -0.0000

Data Index: 687241 (C4 (Web Text))

Max Activating Token Index: 766

Text #17

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.5380. Min Act: -0.0001

Data Index: 165643 (C4 (Web Text))

Max Activating Token Index: 52

Text #18

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.5355. Min Act: -0.0001

Data Index: 646968 (C4 (Web Text))

Max Activating Token Index: 203

Text #19

Max Range: 3.7430. Min Range: -3.7430

Max Act: 2.3882. Min Act: -0.0000

Data Index: 1068248 (C4 (Web Text))

Max Activating Token Index: 912