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

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

Neuron 258 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.7682. Min Act: -0.1700

Data Index: 1395908 (Python Code)

Max Activating Token Index: 326

Text #1

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.5495. Min Act: -0.1700

Data Index: 1549794 (Python Code)

Max Activating Token Index: 553

Text #2

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.5807. Min Act: -0.1700

Data Index: 1597949 (Python Code)

Max Activating Token Index: 311

Text #3

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.5182. Min Act: -0.1700

Data Index: 1375224 (Python Code)

Max Activating Token Index: 988

Text #4

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.4557. Min Act: -0.1700

Data Index: 1638283 (Python Code)

Max Activating Token Index: 98

Text #5

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.4870. Min Act: -0.1699

Data Index: 1437137 (Python Code)

Max Activating Token Index: 609

Text #6

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.5182. Min Act: -0.1700

Data Index: 1470941 (Python Code)

Max Activating Token Index: 362

Text #7

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.4245. Min Act: -0.1700

Data Index: 1387691 (Python Code)

Max Activating Token Index: 780

Text #8

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.4870. Min Act: -0.1700

Data Index: 1618416 (Python Code)

Max Activating Token Index: 158

Text #9

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.4557. Min Act: -0.1700

Data Index: 1516469 (Python Code)

Max Activating Token Index: 331

Text #10

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.4245. Min Act: -0.1683

Data Index: 490137 (C4 (Web Text))

Max Activating Token Index: 928

Text #11

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.3932. Min Act: -0.1700

Data Index: 1368453 (Python Code)

Max Activating Token Index: 48

Text #12

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.3932. Min Act: -0.1699

Data Index: 1430568 (Python Code)

Max Activating Token Index: 1005

Text #13

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.4557. Min Act: -0.1700

Data Index: 1452183 (Python Code)

Max Activating Token Index: 311

Text #14

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.3932. Min Act: -0.1697

Data Index: 644292 (C4 (Web Text))

Max Activating Token Index: 266

Text #15

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.3620. Min Act: -0.1700

Data Index: 1540906 (Python Code)

Max Activating Token Index: 299

Text #16

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.3932. Min Act: -0.1700

Data Index: 1522992 (Python Code)

Max Activating Token Index: 54

Text #17

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.4245. Min Act: -0.1700

Data Index: 1489289 (Python Code)

Max Activating Token Index: 223

Text #18

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.4245. Min Act: -0.1700

Data Index: 1583825 (Python Code)

Max Activating Token Index: 738

Text #19

Max Range: 5.7682. Min Range: -5.7682

Max Act: 5.3932. Min Act: -0.1700

Data Index: 1586652 (Python Code)

Max Activating Token Index: 457