Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4807 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.8303. Min Act: -0.0000

Data Index: 1611128 (Python Code)

Max Activating Token Index: 109

Text #1

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.7485. Min Act: -0.0000

Data Index: 1121033 (C4 (Web Text))

Max Activating Token Index: 920

Text #2

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.6391. Min Act: -0.0000

Data Index: 1513815 (Python Code)

Max Activating Token Index: 629

Text #3

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.6459. Min Act: -0.0000

Data Index: 185886 (C4 (Web Text))

Max Activating Token Index: 342

Text #4

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.5818. Min Act: -0.0000

Data Index: 383092 (C4 (Web Text))

Max Activating Token Index: 336

Text #5

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.5868. Min Act: -0.0000

Data Index: 1453495 (Python Code)

Max Activating Token Index: 237

Text #6

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.5413. Min Act: -0.0000

Data Index: 474613 (C4 (Web Text))

Max Activating Token Index: 71

Text #7

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.5376. Min Act: -0.0000

Data Index: 1282797 (C4 (Web Text))

Max Activating Token Index: 827

Text #8

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4795. Min Act: -0.0000

Data Index: 71487 (C4 (Web Text))

Max Activating Token Index: 649

Text #9

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4709. Min Act: -0.0000

Data Index: 859056 (C4 (Web Text))

Max Activating Token Index: 552

Text #10

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4677. Min Act: -0.0000

Data Index: 762839 (C4 (Web Text))

Max Activating Token Index: 133

Text #11

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4842. Min Act: -0.0000

Data Index: 264757 (C4 (Web Text))

Max Activating Token Index: 1015

Text #12

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4673. Min Act: -0.0000

Data Index: 1504435 (Python Code)

Max Activating Token Index: 847

Text #13

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4704. Min Act: -0.0000

Data Index: 1592566 (Python Code)

Max Activating Token Index: 108

Text #14

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4391. Min Act: -0.0000

Data Index: 1573025 (Python Code)

Max Activating Token Index: 522

Text #15

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4499. Min Act: -0.0000

Data Index: 750026 (C4 (Web Text))

Max Activating Token Index: 692

Text #16

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4772. Min Act: -0.0000

Data Index: 705686 (C4 (Web Text))

Max Activating Token Index: 660

Text #17

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4369. Min Act: -0.0000

Data Index: 1177707 (C4 (Web Text))

Max Activating Token Index: 161

Text #18

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4607. Min Act: -0.0000

Data Index: 1339270 (C4 (Web Text))

Max Activating Token Index: 239

Text #19

Max Range: 0.8303. Min Range: -0.8303

Max Act: 0.4741. Min Act: -0.0000

Data Index: 654164 (C4 (Web Text))

Max Activating Token Index: 980