Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 1418 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 3.2032. Min Range: -3.2032

Max Act: 3.2032. Min Act: -0.1700

Data Index: 1417708 (Python Code)

Max Activating Token Index: 583

Text #1

Max Range: 3.2032. Min Range: -3.2032

Max Act: 3.0755. Min Act: -0.1695

Data Index: 1483131 (Python Code)

Max Activating Token Index: 540

Text #2

Max Range: 3.2032. Min Range: -3.2032

Max Act: 3.0214. Min Act: -0.1700

Data Index: 1430304 (Python Code)

Max Activating Token Index: 909

Text #3

Max Range: 3.2032. Min Range: -3.2032

Max Act: 3.0145. Min Act: -0.1700

Data Index: 1597365 (Python Code)

Max Activating Token Index: 252

Text #4

Max Range: 3.2032. Min Range: -3.2032

Max Act: 3.0072. Min Act: -0.1690

Data Index: 1386559 (Python Code)

Max Activating Token Index: 893

Text #5

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9978. Min Act: -0.1316

Data Index: 1573656 (Python Code)

Max Activating Token Index: 255

Text #6

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9853. Min Act: -0.1680

Data Index: 1524487 (Python Code)

Max Activating Token Index: 679

Text #7

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9982. Min Act: -0.1699

Data Index: 1534396 (Python Code)

Max Activating Token Index: 504

Text #8

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9716. Min Act: -0.1700

Data Index: 680703 (C4 (Web Text))

Max Activating Token Index: 253

Text #9

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9564. Min Act: -0.1700

Data Index: 1375941 (Python Code)

Max Activating Token Index: 140

Text #10

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9676. Min Act: -0.1668

Data Index: 1647111 (Python Code)

Max Activating Token Index: 284

Text #11

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9420. Min Act: -0.1680

Data Index: 1610834 (Python Code)

Max Activating Token Index: 418

Text #12

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9425. Min Act: -0.1694

Data Index: 1620153 (Python Code)

Max Activating Token Index: 635

Text #13

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9191. Min Act: -0.1700

Data Index: 893057 (C4 (Web Text))

Max Activating Token Index: 806

Text #14

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9214. Min Act: -0.1691

Data Index: 1485160 (Python Code)

Max Activating Token Index: 595

Text #15

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9223. Min Act: -0.1687

Data Index: 1636870 (Python Code)

Max Activating Token Index: 308

Text #16

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9106. Min Act: -0.1698

Data Index: 1439262 (Python Code)

Max Activating Token Index: 615

Text #17

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.9022. Min Act: -0.1698

Data Index: 1412090 (Python Code)

Max Activating Token Index: 886

Text #18

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.8871. Min Act: -0.1700

Data Index: 1499872 (Python Code)

Max Activating Token Index: 399

Text #19

Max Range: 3.2032. Min Range: -3.2032

Max Act: 2.8808. Min Act: -0.1700

Data Index: 1520420 (Python Code)

Max Activating Token Index: 480