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

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

Neuron 4000 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.5464. Min Act: -0.0002

Data Index: 212698 (C4 (Web Text))

Max Activating Token Index: 78

Text #1

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.4828. Min Act: -0.0002

Data Index: 12928 (C4 (Web Text))

Max Activating Token Index: 785

Text #2

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.3835. Min Act: -0.0002

Data Index: 1021888 (C4 (Web Text))

Max Activating Token Index: 74

Text #3

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.3369. Min Act: -0.0002

Data Index: 433202 (C4 (Web Text))

Max Activating Token Index: 729

Text #4

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.1924. Min Act: -0.0002

Data Index: 311912 (C4 (Web Text))

Max Activating Token Index: 569

Text #5

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.1972. Min Act: -0.0002

Data Index: 144589 (C4 (Web Text))

Max Activating Token Index: 369

Text #6

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.2042. Min Act: -0.0002

Data Index: 645387 (C4 (Web Text))

Max Activating Token Index: 37

Text #7

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.0876. Min Act: -0.0002

Data Index: 444063 (C4 (Web Text))

Max Activating Token Index: 395

Text #8

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.1173. Min Act: -0.0002

Data Index: 1153855 (C4 (Web Text))

Max Activating Token Index: 209

Text #9

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.0817. Min Act: -0.0002

Data Index: 1211907 (C4 (Web Text))

Max Activating Token Index: 837

Text #10

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.0983. Min Act: -0.0002

Data Index: 727176 (C4 (Web Text))

Max Activating Token Index: 257

Text #11

Max Range: 2.5464. Min Range: -2.5464

Max Act: 1.9960. Min Act: -0.0002

Data Index: 1272702 (C4 (Web Text))

Max Activating Token Index: 810

Text #12

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.0559. Min Act: -0.0002

Data Index: 760954 (C4 (Web Text))

Max Activating Token Index: 675

Text #13

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.0483. Min Act: -0.0002

Data Index: 881781 (C4 (Web Text))

Max Activating Token Index: 87

Text #14

Max Range: 2.5464. Min Range: -2.5464

Max Act: 1.9905. Min Act: -0.0002

Data Index: 323112 (C4 (Web Text))

Max Activating Token Index: 459

Text #15

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.0293. Min Act: -0.0002

Data Index: 220195 (C4 (Web Text))

Max Activating Token Index: 77

Text #16

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.0919. Min Act: -0.0002

Data Index: 177600 (C4 (Web Text))

Max Activating Token Index: 362

Text #17

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.0572. Min Act: -0.0002

Data Index: 666790 (C4 (Web Text))

Max Activating Token Index: 350

Text #18

Max Range: 2.5464. Min Range: -2.5464

Max Act: 1.9783. Min Act: -0.0002

Data Index: 617075 (C4 (Web Text))

Max Activating Token Index: 298

Text #19

Max Range: 2.5464. Min Range: -2.5464

Max Act: 2.0125. Min Act: -0.0002

Data Index: 109801 (C4 (Web Text))

Max Activating Token Index: 903