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

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

Neuron 4954 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.5250. Min Range: -4.5250

Max Act: 4.5250. Min Act: -0.0000

Data Index: 384535 (C4 (Web Text))

Max Activating Token Index: 1020

Text #1

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.7339. Min Act: -0.0000

Data Index: 1393231 (Python Code)

Max Activating Token Index: 259

Text #2

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.8862. Min Act: -0.0000

Data Index: 1523484 (Python Code)

Max Activating Token Index: 506

Text #3

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.4264. Min Act: -0.0000

Data Index: 1589428 (Python Code)

Max Activating Token Index: 756

Text #4

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.2434. Min Act: -0.0000

Data Index: 1632928 (Python Code)

Max Activating Token Index: 496

Text #5

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.6168. Min Act: -0.0000

Data Index: 878219 (C4 (Web Text))

Max Activating Token Index: 724

Text #6

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.3021. Min Act: -0.0000

Data Index: 1464423 (Python Code)

Max Activating Token Index: 325

Text #7

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.1778. Min Act: -0.0000

Data Index: 1454949 (Python Code)

Max Activating Token Index: 388

Text #8

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.0028. Min Act: -0.0000

Data Index: 1564090 (Python Code)

Max Activating Token Index: 1002

Text #9

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.2777. Min Act: -0.0000

Data Index: 1437428 (Python Code)

Max Activating Token Index: 520

Text #10

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.1202. Min Act: -0.0000

Data Index: 1387539 (Python Code)

Max Activating Token Index: 222

Text #11

Max Range: 4.5250. Min Range: -4.5250

Max Act: 3.0209. Min Act: -0.0000

Data Index: 1469380 (Python Code)

Max Activating Token Index: 284

Text #12

Max Range: 4.5250. Min Range: -4.5250

Max Act: 2.9654. Min Act: -0.0000

Data Index: 847215 (C4 (Web Text))

Max Activating Token Index: 689

Text #13

Max Range: 4.5250. Min Range: -4.5250

Max Act: 2.5883. Min Act: -0.0000

Data Index: 1651282 (Python Code)

Max Activating Token Index: 400

Text #14

Max Range: 4.5250. Min Range: -4.5250

Max Act: 2.5334. Min Act: -0.0000

Data Index: 1609229 (Python Code)

Max Activating Token Index: 672

Text #15

Max Range: 4.5250. Min Range: -4.5250

Max Act: 2.6956. Min Act: -0.0000

Data Index: 1442156 (Python Code)

Max Activating Token Index: 350

Text #16

Max Range: 4.5250. Min Range: -4.5250

Max Act: 2.5620. Min Act: -0.0000

Data Index: 1560400 (Python Code)

Max Activating Token Index: 38

Text #17

Max Range: 4.5250. Min Range: -4.5250

Max Act: 2.4593. Min Act: -0.0000

Data Index: 1618831 (Python Code)

Max Activating Token Index: 495

Text #18

Max Range: 4.5250. Min Range: -4.5250

Max Act: 2.3915. Min Act: -0.0000

Data Index: 1598811 (Python Code)

Max Activating Token Index: 882

Text #19

Max Range: 4.5250. Min Range: -4.5250

Max Act: 2.3926. Min Act: -0.0000

Data Index: 1404429 (Python Code)

Max Activating Token Index: 393