Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4712 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 4.0358. Min Range: -4.0358

Max Act: 4.0358. Min Act: -0.0000

Data Index: 740803 (C4 (Web Text))

Max Activating Token Index: 140

Text #1

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.8583. Min Act: -0.0000

Data Index: 368399 (C4 (Web Text))

Max Activating Token Index: 243

Text #2

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.4970. Min Act: -0.0000

Data Index: 464739 (C4 (Web Text))

Max Activating Token Index: 719

Text #3

Max Range: 4.0358. Min Range: -4.0358

Max Act: 4.0104. Min Act: -0.0000

Data Index: 1148987 (C4 (Web Text))

Max Activating Token Index: 739

Text #4

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.3787. Min Act: -0.0000

Data Index: 1088873 (C4 (Web Text))

Max Activating Token Index: 124

Text #5

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.4958. Min Act: -0.0000

Data Index: 646463 (C4 (Web Text))

Max Activating Token Index: 494

Text #6

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.4341. Min Act: -0.0000

Data Index: 1274798 (C4 (Web Text))

Max Activating Token Index: 262

Text #7

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.0421. Min Act: -0.0000

Data Index: 1142624 (C4 (Web Text))

Max Activating Token Index: 609

Text #8

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.3519. Min Act: -0.0000

Data Index: 757156 (C4 (Web Text))

Max Activating Token Index: 169

Text #9

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.1557. Min Act: -0.0000

Data Index: 836118 (C4 (Web Text))

Max Activating Token Index: 673

Text #10

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.1493. Min Act: -0.0000

Data Index: 17213 (C4 (Web Text))

Max Activating Token Index: 377

Text #11

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.0865. Min Act: -0.0000

Data Index: 773464 (C4 (Web Text))

Max Activating Token Index: 720

Text #12

Max Range: 4.0358. Min Range: -4.0358

Max Act: 2.8153. Min Act: -0.0000

Data Index: 1358045 (C4 (Web Text))

Max Activating Token Index: 611

Text #13

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.1683. Min Act: -0.0000

Data Index: 1077060 (C4 (Web Text))

Max Activating Token Index: 176

Text #14

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.1677. Min Act: -0.0000

Data Index: 1280865 (C4 (Web Text))

Max Activating Token Index: 82

Text #15

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.0480. Min Act: -0.0000

Data Index: 397971 (C4 (Web Text))

Max Activating Token Index: 992

Text #16

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.3072. Min Act: -0.0000

Data Index: 1421 (C4 (Web Text))

Max Activating Token Index: 788

Text #17

Max Range: 4.0358. Min Range: -4.0358

Max Act: 2.9978. Min Act: -0.0000

Data Index: 825478 (C4 (Web Text))

Max Activating Token Index: 655

Text #18

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.1256. Min Act: -0.0000

Data Index: 185859 (C4 (Web Text))

Max Activating Token Index: 326

Text #19

Max Range: 4.0358. Min Range: -4.0358

Max Act: 3.2394. Min Act: -0.0000

Data Index: 511047 (C4 (Web Text))

Max Activating Token Index: 211