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

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

Neuron 70 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: 3.6895. Min Range: -3.6895

Max Act: 3.6895. Min Act: -0.1700

Data Index: 1505412 (Python Code)

Max Activating Token Index: 11

Text #1

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.6269. Min Act: -0.1700

Data Index: 509049 (C4 (Web Text))

Max Activating Token Index: 951

Text #2

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5956. Min Act: -0.1700

Data Index: 1276422 (C4 (Web Text))

Max Activating Token Index: 711

Text #3

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.6113. Min Act: -0.1700

Data Index: 1370427 (Python Code)

Max Activating Token Index: 776

Text #4

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.6113. Min Act: -0.1700

Data Index: 456358 (C4 (Web Text))

Max Activating Token Index: 652

Text #5

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5956. Min Act: -0.1700

Data Index: 1442051 (Python Code)

Max Activating Token Index: 638

Text #6

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5956. Min Act: -0.1700

Data Index: 930146 (C4 (Web Text))

Max Activating Token Index: 853

Text #7

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.6113. Min Act: -0.1700

Data Index: 1127288 (C4 (Web Text))

Max Activating Token Index: 589

Text #8

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5643. Min Act: -0.1700

Data Index: 1293827 (C4 (Web Text))

Max Activating Token Index: 81

Text #9

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5643. Min Act: -0.1700

Data Index: 773956 (C4 (Web Text))

Max Activating Token Index: 177

Text #10

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5643. Min Act: -0.1700

Data Index: 839214 (C4 (Web Text))

Max Activating Token Index: 105

Text #11

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5799. Min Act: -0.1700

Data Index: 858931 (C4 (Web Text))

Max Activating Token Index: 841

Text #12

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5643. Min Act: -0.1700

Data Index: 460564 (C4 (Web Text))

Max Activating Token Index: 245

Text #13

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5799. Min Act: -0.1700

Data Index: 1162381 (C4 (Web Text))

Max Activating Token Index: 951

Text #14

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5799. Min Act: -0.1700

Data Index: 707883 (C4 (Web Text))

Max Activating Token Index: 900

Text #15

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5643. Min Act: -0.1700

Data Index: 1163080 (C4 (Web Text))

Max Activating Token Index: 285

Text #16

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5643. Min Act: -0.1700

Data Index: 211652 (C4 (Web Text))

Max Activating Token Index: 624

Text #17

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5643. Min Act: -0.1700

Data Index: 643467 (C4 (Web Text))

Max Activating Token Index: 841

Text #18

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5643. Min Act: -0.1700

Data Index: 607559 (C4 (Web Text))

Max Activating Token Index: 655

Text #19

Max Range: 3.6895. Min Range: -3.6895

Max Act: 3.5486. Min Act: -0.1700

Data Index: 994927 (C4 (Web Text))

Max Activating Token Index: 850