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

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

Neuron 2515 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2738. Min Act: -0.0000

Data Index: 516075 (C4 (Web Text))

Max Activating Token Index: 641

Text #1

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2555. Min Act: -0.0000

Data Index: 986150 (C4 (Web Text))

Max Activating Token Index: 931

Text #2

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.3140. Min Act: -0.0000

Data Index: 732088 (C4 (Web Text))

Max Activating Token Index: 725

Text #3

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2513. Min Act: -0.0000

Data Index: 981332 (C4 (Web Text))

Max Activating Token Index: 539

Text #4

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2383. Min Act: -0.0000

Data Index: 665295 (C4 (Web Text))

Max Activating Token Index: 756

Text #5

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2752. Min Act: -0.0000

Data Index: 597936 (C4 (Web Text))

Max Activating Token Index: 300

Text #6

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2494. Min Act: -0.0000

Data Index: 1222996 (C4 (Web Text))

Max Activating Token Index: 191

Text #7

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2551. Min Act: -0.0000

Data Index: 1521352 (Python Code)

Max Activating Token Index: 485

Text #8

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2442. Min Act: -0.0000

Data Index: 1067131 (C4 (Web Text))

Max Activating Token Index: 386

Text #9

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2564. Min Act: -0.0000

Data Index: 819209 (C4 (Web Text))

Max Activating Token Index: 449

Text #10

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2561. Min Act: -0.0000

Data Index: 504193 (C4 (Web Text))

Max Activating Token Index: 1006

Text #11

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2424. Min Act: -0.0000

Data Index: 342022 (C4 (Web Text))

Max Activating Token Index: 908

Text #12

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2257. Min Act: -0.0000

Data Index: 370508 (C4 (Web Text))

Max Activating Token Index: 523

Text #13

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2340. Min Act: -0.0000

Data Index: 1317724 (C4 (Web Text))

Max Activating Token Index: 695

Text #14

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2771. Min Act: -0.0000

Data Index: 1352401 (C4 (Web Text))

Max Activating Token Index: 305

Text #15

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2013. Min Act: -0.0000

Data Index: 1430634 (Python Code)

Max Activating Token Index: 242

Text #16

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2468. Min Act: -0.0000

Data Index: 1234009 (C4 (Web Text))

Max Activating Token Index: 767

Text #17

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2153. Min Act: -0.0000

Data Index: 336965 (C4 (Web Text))

Max Activating Token Index: 468

Text #18

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2183. Min Act: -0.0000

Data Index: 1271940 (C4 (Web Text))

Max Activating Token Index: 256

Text #19

Max Range: 0.3140. Min Range: -0.3140

Max Act: 0.2321. Min Act: -0.0000

Data Index: 1137410 (C4 (Web Text))

Max Activating Token Index: 1020