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

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

Neuron 130 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: 3.5290. Min Range: -3.5290

Max Act: 3.5133. Min Act: -0.1700

Data Index: 1484129 (Python Code)

Max Activating Token Index: 363

Text #1

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.5290. Min Act: -0.1700

Data Index: 1370292 (Python Code)

Max Activating Token Index: 599

Text #2

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.4820. Min Act: -0.1700

Data Index: 1512035 (Python Code)

Max Activating Token Index: 694

Text #3

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.4820. Min Act: -0.1700

Data Index: 1408943 (Python Code)

Max Activating Token Index: 756

Text #4

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3722. Min Act: -0.1700

Data Index: 1367840 (Python Code)

Max Activating Token Index: 555

Text #5

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3879. Min Act: -0.1700

Data Index: 1441811 (Python Code)

Max Activating Token Index: 344

Text #6

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3565. Min Act: -0.1700

Data Index: 1005543 (C4 (Web Text))

Max Activating Token Index: 640

Text #7

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3722. Min Act: -0.1700

Data Index: 1517111 (Python Code)

Max Activating Token Index: 895

Text #8

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3408. Min Act: -0.1700

Data Index: 1046572 (C4 (Web Text))

Max Activating Token Index: 85

Text #9

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3722. Min Act: -0.1700

Data Index: 1561831 (Python Code)

Max Activating Token Index: 578

Text #10

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3251. Min Act: -0.1700

Data Index: 1369833 (Python Code)

Max Activating Token Index: 681

Text #11

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3408. Min Act: -0.1700

Data Index: 1602636 (Python Code)

Max Activating Token Index: 36

Text #12

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3094. Min Act: -0.1700

Data Index: 1426269 (Python Code)

Max Activating Token Index: 391

Text #13

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3251. Min Act: -0.1700

Data Index: 1629998 (Python Code)

Max Activating Token Index: 554

Text #14

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3251. Min Act: -0.1700

Data Index: 1615442 (Python Code)

Max Activating Token Index: 213

Text #15

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.2937. Min Act: -0.1700

Data Index: 1371895 (Python Code)

Max Activating Token Index: 114

Text #16

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3094. Min Act: -0.1700

Data Index: 1261418 (C4 (Web Text))

Max Activating Token Index: 466

Text #17

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3094. Min Act: -0.1700

Data Index: 1534163 (Python Code)

Max Activating Token Index: 195

Text #18

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.3094. Min Act: -0.1700

Data Index: 1571675 (Python Code)

Max Activating Token Index: 391

Text #19

Max Range: 3.5290. Min Range: -3.5290

Max Act: 3.2937. Min Act: -0.1700

Data Index: 1479046 (Python Code)

Max Activating Token Index: 740