Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 284 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0633. Min Act: -0.0006

Data Index: 1147920 (C4 (Web Text))

Max Activating Token Index: 654

Text #1

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0646. Min Act: -0.0005

Data Index: 158839 (C4 (Web Text))

Max Activating Token Index: 244

Text #2

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0595. Min Act: -0.0006

Data Index: 77837 (C4 (Web Text))

Max Activating Token Index: 880

Text #3

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0579. Min Act: -0.0006

Data Index: 438270 (C4 (Web Text))

Max Activating Token Index: 325

Text #4

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0571. Min Act: -0.0006

Data Index: 615690 (C4 (Web Text))

Max Activating Token Index: 81

Text #5

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0546. Min Act: -0.0006

Data Index: 1279843 (C4 (Web Text))

Max Activating Token Index: 316

Text #6

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0538. Min Act: -0.0006

Data Index: 112860 (C4 (Web Text))

Max Activating Token Index: 78

Text #7

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0544. Min Act: -0.0005

Data Index: 980246 (C4 (Web Text))

Max Activating Token Index: 34

Text #8

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0540. Min Act: -0.0006

Data Index: 1466225 (Python Code)

Max Activating Token Index: 95

Text #9

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0551. Min Act: -0.0006

Data Index: 823774 (C4 (Web Text))

Max Activating Token Index: 636

Text #10

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0548. Min Act: -0.0006

Data Index: 786701 (C4 (Web Text))

Max Activating Token Index: 1004

Text #11

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0540. Min Act: -0.0005

Data Index: 247172 (C4 (Web Text))

Max Activating Token Index: 27

Text #12

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0517. Min Act: -0.0006

Data Index: 505403 (C4 (Web Text))

Max Activating Token Index: 916

Text #13

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0547. Min Act: -0.0006

Data Index: 1204923 (C4 (Web Text))

Max Activating Token Index: 361

Text #14

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0516. Min Act: -0.0006

Data Index: 896491 (C4 (Web Text))

Max Activating Token Index: 469

Text #15

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0490. Min Act: -0.0005

Data Index: 532139 (C4 (Web Text))

Max Activating Token Index: 53

Text #16

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0522. Min Act: -0.0006

Data Index: 258171 (C4 (Web Text))

Max Activating Token Index: 440

Text #17

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0527. Min Act: -0.0006

Data Index: 563637 (C4 (Web Text))

Max Activating Token Index: 798

Text #18

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0537. Min Act: -0.0006

Data Index: 642980 (C4 (Web Text))

Max Activating Token Index: 443

Text #19

Max Range: 0.0646. Min Range: -0.0646

Max Act: 0.0500. Min Act: -0.0006

Data Index: 362698 (C4 (Web Text))

Max Activating Token Index: 877