Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 393 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.2495. Min Range: -5.2495

Max Act: 5.2495. Min Act: -0.1700

Data Index: 761140 (C4 (Web Text))

Max Activating Token Index: 882

Text #1

Max Range: 5.2495. Min Range: -5.2495

Max Act: 5.0620. Min Act: -0.1700

Data Index: 446474 (C4 (Web Text))

Max Activating Token Index: 543

Text #2

Max Range: 5.2495. Min Range: -5.2495

Max Act: 5.0307. Min Act: -0.1700

Data Index: 740292 (C4 (Web Text))

Max Activating Token Index: 748

Text #3

Max Range: 5.2495. Min Range: -5.2495

Max Act: 5.0307. Min Act: -0.1700

Data Index: 640872 (C4 (Web Text))

Max Activating Token Index: 522

Text #4

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.9370. Min Act: -0.1700

Data Index: 586903 (C4 (Web Text))

Max Activating Token Index: 252

Text #5

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.9057. Min Act: -0.1700

Data Index: 718445 (C4 (Web Text))

Max Activating Token Index: 423

Text #6

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8745. Min Act: -0.1700

Data Index: 46407 (C4 (Web Text))

Max Activating Token Index: 131

Text #7

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8745. Min Act: -0.1700

Data Index: 1226363 (C4 (Web Text))

Max Activating Token Index: 135

Text #8

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8120. Min Act: -0.1700

Data Index: 1597655 (Python Code)

Max Activating Token Index: 949

Text #9

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8432. Min Act: -0.1700

Data Index: 289936 (C4 (Web Text))

Max Activating Token Index: 370

Text #10

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8745. Min Act: -0.1700

Data Index: 506800 (C4 (Web Text))

Max Activating Token Index: 58

Text #11

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8120. Min Act: -0.1700

Data Index: 988732 (C4 (Web Text))

Max Activating Token Index: 989

Text #12

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8432. Min Act: -0.1700

Data Index: 1535017 (Python Code)

Max Activating Token Index: 60

Text #13

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8120. Min Act: -0.1699

Data Index: 1171137 (C4 (Web Text))

Max Activating Token Index: 437

Text #14

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.7495. Min Act: -0.1700

Data Index: 763509 (C4 (Web Text))

Max Activating Token Index: 751

Text #15

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8745. Min Act: -0.1700

Data Index: 1282010 (C4 (Web Text))

Max Activating Token Index: 398

Text #16

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8432. Min Act: -0.1700

Data Index: 395440 (C4 (Web Text))

Max Activating Token Index: 501

Text #17

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.8120. Min Act: -0.1700

Data Index: 576728 (C4 (Web Text))

Max Activating Token Index: 56

Text #18

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.6557. Min Act: -0.1700

Data Index: 1530849 (Python Code)

Max Activating Token Index: 845

Text #19

Max Range: 5.2495. Min Range: -5.2495

Max Act: 4.7807. Min Act: -0.1700

Data Index: 1539235 (Python Code)

Max Activating Token Index: 551