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

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

Neuron 859 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: 7.8614. Min Range: -7.8614

Max Act: 7.8301. Min Act: -0.1700

Data Index: 414678 (C4 (Web Text))

Max Activating Token Index: 231

Text #1

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.8614. Min Act: -0.1700

Data Index: 24772 (C4 (Web Text))

Max Activating Token Index: 434

Text #2

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.8301. Min Act: -0.1700

Data Index: 1542799 (Python Code)

Max Activating Token Index: 367

Text #3

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.7676. Min Act: -0.1700

Data Index: 12703 (C4 (Web Text))

Max Activating Token Index: 506

Text #4

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.7364. Min Act: -0.1700

Data Index: 1043458 (C4 (Web Text))

Max Activating Token Index: 898

Text #5

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.7051. Min Act: -0.1700

Data Index: 1564769 (Python Code)

Max Activating Token Index: 556

Text #6

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.7364. Min Act: -0.1700

Data Index: 54802 (C4 (Web Text))

Max Activating Token Index: 125

Text #7

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.6739. Min Act: -0.1700

Data Index: 1565580 (Python Code)

Max Activating Token Index: 363

Text #8

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.6426. Min Act: -0.1700

Data Index: 1466298 (Python Code)

Max Activating Token Index: 334

Text #9

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.5176. Min Act: -0.1700

Data Index: 1405565 (Python Code)

Max Activating Token Index: 797

Text #10

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.4551. Min Act: -0.1700

Data Index: 1412973 (Python Code)

Max Activating Token Index: 221

Text #11

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.5489. Min Act: -0.1700

Data Index: 1321864 (C4 (Web Text))

Max Activating Token Index: 781

Text #12

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.5801. Min Act: -0.1700

Data Index: 1135500 (C4 (Web Text))

Max Activating Token Index: 343

Text #13

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.6114. Min Act: -0.1700

Data Index: 392138 (C4 (Web Text))

Max Activating Token Index: 814

Text #14

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.5176. Min Act: -0.1700

Data Index: 145242 (C4 (Web Text))

Max Activating Token Index: 387

Text #15

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.5176. Min Act: -0.1700

Data Index: 1047066 (C4 (Web Text))

Max Activating Token Index: 1020

Text #16

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.4551. Min Act: -0.1700

Data Index: 583138 (C4 (Web Text))

Max Activating Token Index: 869

Text #17

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.4864. Min Act: -0.1700

Data Index: 685553 (C4 (Web Text))

Max Activating Token Index: 50

Text #18

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.4239. Min Act: -0.1700

Data Index: 1394185 (Python Code)

Max Activating Token Index: 935

Text #19

Max Range: 7.8614. Min Range: -7.8614

Max Act: 7.4239. Min Act: -0.1700

Data Index: 1508314 (Python Code)

Max Activating Token Index: 385