Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 946 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.2084. Min Act: -0.1700

Data Index: 1243485 (C4 (Web Text))

Max Activating Token Index: 928

Text #1

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.1611. Min Act: -0.1700

Data Index: 286094 (C4 (Web Text))

Max Activating Token Index: 559

Text #2

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.1454. Min Act: -0.1700

Data Index: 360160 (C4 (Web Text))

Max Activating Token Index: 396

Text #3

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.1296. Min Act: -0.1700

Data Index: 1330901 (C4 (Web Text))

Max Activating Token Index: 976

Text #4

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.1296. Min Act: -0.1700

Data Index: 900327 (C4 (Web Text))

Max Activating Token Index: 577

Text #5

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.1139. Min Act: -0.1700

Data Index: 429828 (C4 (Web Text))

Max Activating Token Index: 670

Text #6

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.1139. Min Act: -0.1700

Data Index: 45554 (C4 (Web Text))

Max Activating Token Index: 309

Text #7

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0823. Min Act: -0.1700

Data Index: 1640467 (Python Code)

Max Activating Token Index: 190

Text #8

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0823. Min Act: -0.1700

Data Index: 456183 (C4 (Web Text))

Max Activating Token Index: 287

Text #9

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0823. Min Act: -0.1700

Data Index: 804351 (C4 (Web Text))

Max Activating Token Index: 244

Text #10

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0666. Min Act: -0.1700

Data Index: 130912 (C4 (Web Text))

Max Activating Token Index: 268

Text #11

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0666. Min Act: -0.1700

Data Index: 1505869 (Python Code)

Max Activating Token Index: 49

Text #12

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0666. Min Act: -0.1695

Data Index: 1596222 (Python Code)

Max Activating Token Index: 927

Text #13

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0508. Min Act: -0.1700

Data Index: 144494 (C4 (Web Text))

Max Activating Token Index: 148

Text #14

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0508. Min Act: -0.1700

Data Index: 1466733 (Python Code)

Max Activating Token Index: 390

Text #15

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0350. Min Act: -0.1699

Data Index: 1591058 (Python Code)

Max Activating Token Index: 833

Text #16

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0034. Min Act: -0.1700

Data Index: 1161686 (C4 (Web Text))

Max Activating Token Index: 243

Text #17

Max Range: 3.2084. Min Range: -3.2084

Max Act: 3.0034. Min Act: -0.1700

Data Index: 1020674 (C4 (Web Text))

Max Activating Token Index: 67

Text #18

Max Range: 3.2084. Min Range: -3.2084

Max Act: 2.9876. Min Act: -0.1700

Data Index: 159822 (C4 (Web Text))

Max Activating Token Index: 267

Text #19

Max Range: 3.2084. Min Range: -3.2084

Max Act: 2.9718. Min Act: -0.1700

Data Index: 1532709 (Python Code)

Max Activating Token Index: 972