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

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

Neuron 1619 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.5371. Min Act: -0.1700

Data Index: 1359752 (C4 (Web Text))

Max Activating Token Index: 229

Text #1

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.3018. Min Act: -0.1700

Data Index: 906518 (C4 (Web Text))

Max Activating Token Index: 207

Text #2

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.3175. Min Act: -0.1700

Data Index: 1650633 (Python Code)

Max Activating Token Index: 190

Text #3

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.3018. Min Act: -0.1700

Data Index: 595215 (C4 (Web Text))

Max Activating Token Index: 552

Text #4

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.2861. Min Act: -0.1700

Data Index: 158120 (C4 (Web Text))

Max Activating Token Index: 701

Text #5

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.3018. Min Act: -0.1700

Data Index: 113253 (C4 (Web Text))

Max Activating Token Index: 179

Text #6

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.2547. Min Act: -0.1700

Data Index: 1513952 (Python Code)

Max Activating Token Index: 463

Text #7

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.2547. Min Act: -0.1700

Data Index: 5424 (C4 (Web Text))

Max Activating Token Index: 443

Text #8

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.2075. Min Act: -0.1700

Data Index: 1290964 (C4 (Web Text))

Max Activating Token Index: 749

Text #9

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.2075. Min Act: -0.1700

Data Index: 714617 (C4 (Web Text))

Max Activating Token Index: 649

Text #10

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.2232. Min Act: -0.1700

Data Index: 1425950 (Python Code)

Max Activating Token Index: 886

Text #11

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.2232. Min Act: -0.1700

Data Index: 774174 (C4 (Web Text))

Max Activating Token Index: 98

Text #12

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.1918. Min Act: -0.1700

Data Index: 1561491 (Python Code)

Max Activating Token Index: 335

Text #13

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.1918. Min Act: -0.1700

Data Index: 1654688 (Python Code)

Max Activating Token Index: 736

Text #14

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.1760. Min Act: -0.1700

Data Index: 1631751 (Python Code)

Max Activating Token Index: 640

Text #15

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.1918. Min Act: -0.1700

Data Index: 1508940 (Python Code)

Max Activating Token Index: 736

Text #16

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.1603. Min Act: -0.1700

Data Index: 1371138 (Python Code)

Max Activating Token Index: 806

Text #17

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.1760. Min Act: -0.1700

Data Index: 1453216 (Python Code)

Max Activating Token Index: 826

Text #18

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.1760. Min Act: -0.1700

Data Index: 1480344 (Python Code)

Max Activating Token Index: 229

Text #19

Max Range: 3.5371. Min Range: -3.5371

Max Act: 3.1760. Min Act: -0.1700

Data Index: 1408092 (Python Code)

Max Activating Token Index: 672