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

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

Neuron 1355 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0507. Min Act: -0.0006

Data Index: 1168711 (C4 (Web Text))

Max Activating Token Index: 238

Text #1

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0505. Min Act: -0.0006

Data Index: 1108557 (C4 (Web Text))

Max Activating Token Index: 98

Text #2

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0512. Min Act: -0.0006

Data Index: 550801 (C4 (Web Text))

Max Activating Token Index: 240

Text #3

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0490. Min Act: -0.0006

Data Index: 151990 (C4 (Web Text))

Max Activating Token Index: 36

Text #4

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0435. Min Act: -0.0006

Data Index: 333201 (C4 (Web Text))

Max Activating Token Index: 1003

Text #5

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0462. Min Act: -0.0006

Data Index: 45839 (C4 (Web Text))

Max Activating Token Index: 681

Text #6

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0439. Min Act: -0.0006

Data Index: 891649 (C4 (Web Text))

Max Activating Token Index: 702

Text #7

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0450. Min Act: -0.0006

Data Index: 637716 (C4 (Web Text))

Max Activating Token Index: 323

Text #8

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0436. Min Act: -0.0005

Data Index: 376358 (C4 (Web Text))

Max Activating Token Index: 11

Text #9

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0423. Min Act: -0.0006

Data Index: 1514015 (Python Code)

Max Activating Token Index: 30

Text #10

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0448. Min Act: -0.0005

Data Index: 1127832 (C4 (Web Text))

Max Activating Token Index: 609

Text #11

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0430. Min Act: -0.0006

Data Index: 997058 (C4 (Web Text))

Max Activating Token Index: 250

Text #12

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0446. Min Act: -0.0006

Data Index: 315618 (C4 (Web Text))

Max Activating Token Index: 29

Text #13

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0433. Min Act: -0.0006

Data Index: 565312 (C4 (Web Text))

Max Activating Token Index: 599

Text #14

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0426. Min Act: -0.0006

Data Index: 341267 (C4 (Web Text))

Max Activating Token Index: 937

Text #15

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0418. Min Act: -0.0006

Data Index: 600414 (C4 (Web Text))

Max Activating Token Index: 655

Text #16

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0433. Min Act: -0.0006

Data Index: 188593 (C4 (Web Text))

Max Activating Token Index: 724

Text #17

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0413. Min Act: -0.0006

Data Index: 1259543 (C4 (Web Text))

Max Activating Token Index: 117

Text #18

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0410. Min Act: -0.0006

Data Index: 589562 (C4 (Web Text))

Max Activating Token Index: 848

Text #19

Max Range: 0.0512. Min Range: -0.0512

Max Act: 0.0391. Min Act: -0.0006

Data Index: 994333 (C4 (Web Text))

Max Activating Token Index: 218