Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 87 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.4601. Min Act: -0.1700

Data Index: 228591 (C4 (Web Text))

Max Activating Token Index: 975

Text #1

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3817. Min Act: -0.1700

Data Index: 306194 (C4 (Web Text))

Max Activating Token Index: 746

Text #2

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3660. Min Act: -0.1700

Data Index: 181820 (C4 (Web Text))

Max Activating Token Index: 397

Text #3

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3660. Min Act: -0.1700

Data Index: 77725 (C4 (Web Text))

Max Activating Token Index: 514

Text #4

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3346. Min Act: -0.1700

Data Index: 1341106 (C4 (Web Text))

Max Activating Token Index: 309

Text #5

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3503. Min Act: -0.1700

Data Index: 1102562 (C4 (Web Text))

Max Activating Token Index: 616

Text #6

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3189. Min Act: -0.1700

Data Index: 691505 (C4 (Web Text))

Max Activating Token Index: 936

Text #7

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3032. Min Act: -0.1700

Data Index: 516655 (C4 (Web Text))

Max Activating Token Index: 19

Text #8

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3032. Min Act: -0.1700

Data Index: 361021 (C4 (Web Text))

Max Activating Token Index: 738

Text #9

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3032. Min Act: -0.1700

Data Index: 1295790 (C4 (Web Text))

Max Activating Token Index: 612

Text #10

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.2875. Min Act: -0.1700

Data Index: 1002736 (C4 (Web Text))

Max Activating Token Index: 621

Text #11

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3032. Min Act: -0.1700

Data Index: 1039662 (C4 (Web Text))

Max Activating Token Index: 737

Text #12

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.2718. Min Act: -0.1700

Data Index: 1433845 (Python Code)

Max Activating Token Index: 650

Text #13

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.3032. Min Act: -0.1700

Data Index: 670121 (C4 (Web Text))

Max Activating Token Index: 844

Text #14

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.2875. Min Act: -0.1700

Data Index: 1267001 (C4 (Web Text))

Max Activating Token Index: 87

Text #15

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.2718. Min Act: -0.1700

Data Index: 1262112 (C4 (Web Text))

Max Activating Token Index: 446

Text #16

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.2718. Min Act: -0.1700

Data Index: 723552 (C4 (Web Text))

Max Activating Token Index: 34

Text #17

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.2875. Min Act: -0.1700

Data Index: 848289 (C4 (Web Text))

Max Activating Token Index: 786

Text #18

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.2875. Min Act: -0.1700

Data Index: 457914 (C4 (Web Text))

Max Activating Token Index: 226

Text #19

Max Range: 3.4601. Min Range: -3.4601

Max Act: 3.2875. Min Act: -0.1700

Data Index: 824859 (C4 (Web Text))

Max Activating Token Index: 732