Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 999 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0607. Min Act: -0.0004

Data Index: 49600 (C4 (Web Text))

Max Activating Token Index: 330

Text #1

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0551. Min Act: -0.0004

Data Index: 456185 (C4 (Web Text))

Max Activating Token Index: 621

Text #2

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0586. Min Act: -0.0004

Data Index: 1125770 (C4 (Web Text))

Max Activating Token Index: 370

Text #3

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0526. Min Act: -0.0004

Data Index: 734435 (C4 (Web Text))

Max Activating Token Index: 258

Text #4

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0512. Min Act: -0.0004

Data Index: 926559 (C4 (Web Text))

Max Activating Token Index: 581

Text #5

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0509. Min Act: -0.0004

Data Index: 41540 (C4 (Web Text))

Max Activating Token Index: 183

Text #6

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0492. Min Act: -0.0004

Data Index: 235294 (C4 (Web Text))

Max Activating Token Index: 202

Text #7

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0492. Min Act: -0.0004

Data Index: 406300 (C4 (Web Text))

Max Activating Token Index: 485

Text #8

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0485. Min Act: -0.0004

Data Index: 1001851 (C4 (Web Text))

Max Activating Token Index: 130

Text #9

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0468. Min Act: -0.0004

Data Index: 702030 (C4 (Web Text))

Max Activating Token Index: 585

Text #10

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0483. Min Act: -0.0004

Data Index: 89692 (C4 (Web Text))

Max Activating Token Index: 155

Text #11

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0483. Min Act: -0.0004

Data Index: 571482 (C4 (Web Text))

Max Activating Token Index: 25

Text #12

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0473. Min Act: -0.0004

Data Index: 396888 (C4 (Web Text))

Max Activating Token Index: 31

Text #13

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0481. Min Act: -0.0004

Data Index: 580186 (C4 (Web Text))

Max Activating Token Index: 270

Text #14

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0443. Min Act: -0.0004

Data Index: 224943 (C4 (Web Text))

Max Activating Token Index: 487

Text #15

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0472. Min Act: -0.0004

Data Index: 1164012 (C4 (Web Text))

Max Activating Token Index: 654

Text #16

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0473. Min Act: -0.0004

Data Index: 714307 (C4 (Web Text))

Max Activating Token Index: 263

Text #17

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0453. Min Act: -0.0004

Data Index: 621232 (C4 (Web Text))

Max Activating Token Index: 419

Text #18

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0445. Min Act: -0.0004

Data Index: 1510392 (Python Code)

Max Activating Token Index: 305

Text #19

Max Range: 0.0607. Min Range: -0.0607

Max Act: 0.0467. Min Act: -0.0004

Data Index: 1111293 (C4 (Web Text))

Max Activating Token Index: 443