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

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

Neuron 1120 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: 3.7207. Min Range: -3.7207

Max Act: 3.7207. Min Act: -0.1699

Data Index: 1301243 (C4 (Web Text))

Max Activating Token Index: 498

Text #1

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.5564. Min Act: -0.1700

Data Index: 1572432 (Python Code)

Max Activating Token Index: 506

Text #2

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.5651. Min Act: -0.1700

Data Index: 1414342 (Python Code)

Max Activating Token Index: 846

Text #3

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.4948. Min Act: -0.1700

Data Index: 1628146 (Python Code)

Max Activating Token Index: 475

Text #4

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.5029. Min Act: -0.1700

Data Index: 1360246 (Python Code)

Max Activating Token Index: 337

Text #5

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.4615. Min Act: -0.1699

Data Index: 1645809 (Python Code)

Max Activating Token Index: 805

Text #6

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.4176. Min Act: -0.1669

Data Index: 1160417 (C4 (Web Text))

Max Activating Token Index: 465

Text #7

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3995. Min Act: -0.1698

Data Index: 1655964 (Python Code)

Max Activating Token Index: 669

Text #8

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.4021. Min Act: -0.1699

Data Index: 612092 (C4 (Web Text))

Max Activating Token Index: 213

Text #9

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3926. Min Act: -0.1689

Data Index: 1050922 (C4 (Web Text))

Max Activating Token Index: 821

Text #10

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3520. Min Act: -0.1673

Data Index: 1628687 (Python Code)

Max Activating Token Index: 456

Text #11

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3599. Min Act: -0.1699

Data Index: 1383242 (Python Code)

Max Activating Token Index: 553

Text #12

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3585. Min Act: -0.1652

Data Index: 64987 (C4 (Web Text))

Max Activating Token Index: 670

Text #13

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3242. Min Act: -0.1699

Data Index: 1526077 (Python Code)

Max Activating Token Index: 504

Text #14

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3317. Min Act: -0.1700

Data Index: 1502935 (Python Code)

Max Activating Token Index: 984

Text #15

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3313. Min Act: -0.1690

Data Index: 9877 (C4 (Web Text))

Max Activating Token Index: 623

Text #16

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.2847. Min Act: -0.1700

Data Index: 367911 (C4 (Web Text))

Max Activating Token Index: 859

Text #17

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3039. Min Act: -0.1700

Data Index: 1376817 (Python Code)

Max Activating Token Index: 834

Text #18

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.3214. Min Act: -0.1700

Data Index: 1607760 (Python Code)

Max Activating Token Index: 640

Text #19

Max Range: 3.7207. Min Range: -3.7207

Max Act: 3.2927. Min Act: -0.1686

Data Index: 699788 (C4 (Web Text))

Max Activating Token Index: 253