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

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

Neuron 1928 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.5801. Min Act: -0.1700

Data Index: 1428258 (Python Code)

Max Activating Token Index: 1004

Text #1

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.5174. Min Act: -0.1700

Data Index: 1444712 (Python Code)

Max Activating Token Index: 697

Text #2

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.5174. Min Act: -0.1700

Data Index: 1546401 (Python Code)

Max Activating Token Index: 268

Text #3

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.4861. Min Act: -0.1700

Data Index: 1563977 (Python Code)

Max Activating Token Index: 196

Text #4

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.4547. Min Act: -0.1700

Data Index: 584075 (C4 (Web Text))

Max Activating Token Index: 856

Text #5

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.4234. Min Act: -0.1700

Data Index: 1398133 (Python Code)

Max Activating Token Index: 607

Text #6

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.4547. Min Act: -0.1700

Data Index: 1567303 (Python Code)

Max Activating Token Index: 891

Text #7

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.3606. Min Act: -0.1700

Data Index: 1561987 (Python Code)

Max Activating Token Index: 412

Text #8

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.3449. Min Act: -0.1700

Data Index: 1467104 (Python Code)

Max Activating Token Index: 624

Text #9

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.4234. Min Act: -0.1700

Data Index: 1519556 (Python Code)

Max Activating Token Index: 194

Text #10

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.3606. Min Act: -0.1700

Data Index: 1632863 (Python Code)

Max Activating Token Index: 125

Text #11

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.3606. Min Act: -0.1700

Data Index: 1641248 (Python Code)

Max Activating Token Index: 768

Text #12

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.3920. Min Act: -0.1700

Data Index: 1656164 (Python Code)

Max Activating Token Index: 410

Text #13

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.3292. Min Act: -0.1700

Data Index: 1460138 (Python Code)

Max Activating Token Index: 849

Text #14

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.3135. Min Act: -0.1700

Data Index: 1487042 (Python Code)

Max Activating Token Index: 414

Text #15

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.2821. Min Act: -0.1700

Data Index: 1389028 (Python Code)

Max Activating Token Index: 738

Text #16

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.2821. Min Act: -0.1700

Data Index: 1448565 (Python Code)

Max Activating Token Index: 989

Text #17

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.2507. Min Act: -0.1700

Data Index: 1586507 (Python Code)

Max Activating Token Index: 386

Text #18

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.3292. Min Act: -0.1700

Data Index: 1447529 (Python Code)

Max Activating Token Index: 488

Text #19

Max Range: 3.5801. Min Range: -3.5801

Max Act: 3.3449. Min Act: -0.1700

Data Index: 1383392 (Python Code)

Max Activating Token Index: 916