Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 580 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2841. Min Act: -0.0002

Data Index: 237231 (C4 (Web Text))

Max Activating Token Index: 731

Text #1

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2742. Min Act: -0.0002

Data Index: 193152 (C4 (Web Text))

Max Activating Token Index: 198

Text #2

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2562. Min Act: -0.0002

Data Index: 835294 (C4 (Web Text))

Max Activating Token Index: 169

Text #3

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2615. Min Act: -0.0002

Data Index: 414624 (C4 (Web Text))

Max Activating Token Index: 305

Text #4

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2418. Min Act: -0.0002

Data Index: 482199 (C4 (Web Text))

Max Activating Token Index: 654

Text #5

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2480. Min Act: -0.0002

Data Index: 1236684 (C4 (Web Text))

Max Activating Token Index: 620

Text #6

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2474. Min Act: -0.0002

Data Index: 1336089 (C4 (Web Text))

Max Activating Token Index: 316

Text #7

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2540. Min Act: -0.0002

Data Index: 669433 (C4 (Web Text))

Max Activating Token Index: 187

Text #8

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2467. Min Act: -0.0002

Data Index: 643910 (C4 (Web Text))

Max Activating Token Index: 818

Text #9

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2380. Min Act: -0.0002

Data Index: 991552 (C4 (Web Text))

Max Activating Token Index: 200

Text #10

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2483. Min Act: -0.0002

Data Index: 141666 (C4 (Web Text))

Max Activating Token Index: 135

Text #11

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2391. Min Act: -0.0002

Data Index: 1198767 (C4 (Web Text))

Max Activating Token Index: 561

Text #12

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2488. Min Act: -0.0002

Data Index: 332007 (C4 (Web Text))

Max Activating Token Index: 754

Text #13

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2532. Min Act: -0.0002

Data Index: 547982 (C4 (Web Text))

Max Activating Token Index: 739

Text #14

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2624. Min Act: -0.0002

Data Index: 398174 (C4 (Web Text))

Max Activating Token Index: 1017

Text #15

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2408. Min Act: -0.0002

Data Index: 201308 (C4 (Web Text))

Max Activating Token Index: 602

Text #16

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2343. Min Act: -0.0002

Data Index: 1048063 (C4 (Web Text))

Max Activating Token Index: 75

Text #17

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2447. Min Act: -0.0002

Data Index: 1120531 (C4 (Web Text))

Max Activating Token Index: 673

Text #18

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2323. Min Act: -0.0002

Data Index: 279151 (C4 (Web Text))

Max Activating Token Index: 172

Text #19

Max Range: 0.2841. Min Range: -0.2841

Max Act: 0.2490. Min Act: -0.0002

Data Index: 959271 (C4 (Web Text))

Max Activating Token Index: 684