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

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

Neuron 3339 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.5129. Min Act: -0.0003

Data Index: 571869 (C4 (Web Text))

Max Activating Token Index: 887

Text #1

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.5220. Min Act: -0.0003

Data Index: 1111095 (C4 (Web Text))

Max Activating Token Index: 255

Text #2

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4889. Min Act: -0.0003

Data Index: 941544 (C4 (Web Text))

Max Activating Token Index: 12

Text #3

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.5311. Min Act: -0.0003

Data Index: 1216653 (C4 (Web Text))

Max Activating Token Index: 804

Text #4

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.5032. Min Act: -0.0003

Data Index: 1201346 (C4 (Web Text))

Max Activating Token Index: 8

Text #5

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4741. Min Act: -0.0003

Data Index: 547626 (C4 (Web Text))

Max Activating Token Index: 313

Text #6

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4231. Min Act: -0.0003

Data Index: 170313 (C4 (Web Text))

Max Activating Token Index: 324

Text #7

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4800. Min Act: -0.0003

Data Index: 285750 (C4 (Web Text))

Max Activating Token Index: 19

Text #8

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4761. Min Act: -0.0003

Data Index: 1164676 (C4 (Web Text))

Max Activating Token Index: 38

Text #9

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4558. Min Act: -0.0003

Data Index: 916941 (C4 (Web Text))

Max Activating Token Index: 577

Text #10

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4686. Min Act: -0.0003

Data Index: 1138020 (C4 (Web Text))

Max Activating Token Index: 19

Text #11

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4487. Min Act: -0.0003

Data Index: 579382 (C4 (Web Text))

Max Activating Token Index: 359

Text #12

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4474. Min Act: -0.0003

Data Index: 979618 (C4 (Web Text))

Max Activating Token Index: 299

Text #13

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4606. Min Act: -0.0003

Data Index: 346661 (C4 (Web Text))

Max Activating Token Index: 336

Text #14

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4706. Min Act: -0.0003

Data Index: 820863 (C4 (Web Text))

Max Activating Token Index: 829

Text #15

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4667. Min Act: -0.0003

Data Index: 142290 (C4 (Web Text))

Max Activating Token Index: 617

Text #16

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4387. Min Act: -0.0003

Data Index: 1127237 (C4 (Web Text))

Max Activating Token Index: 459

Text #17

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.3984. Min Act: -0.0003

Data Index: 473105 (C4 (Web Text))

Max Activating Token Index: 679

Text #18

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4515. Min Act: -0.0003

Data Index: 207000 (C4 (Web Text))

Max Activating Token Index: 6

Text #19

Max Range: 0.5311. Min Range: -0.5311

Max Act: 0.4633. Min Act: -0.0003

Data Index: 556988 (C4 (Web Text))

Max Activating Token Index: 439