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

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

Neuron 1422 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: 3.7335. Min Range: -3.7335

Max Act: 3.7335. Min Act: -0.1700

Data Index: 1619050 (Python Code)

Max Activating Token Index: 142

Text #1

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5770. Min Act: -0.1700

Data Index: 1365409 (Python Code)

Max Activating Token Index: 940

Text #2

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5457. Min Act: -0.1700

Data Index: 1374532 (Python Code)

Max Activating Token Index: 139

Text #3

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5457. Min Act: -0.1700

Data Index: 900128 (C4 (Web Text))

Max Activating Token Index: 498

Text #4

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5143. Min Act: -0.1700

Data Index: 1472658 (Python Code)

Max Activating Token Index: 219

Text #5

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5143. Min Act: -0.1700

Data Index: 1583639 (Python Code)

Max Activating Token Index: 987

Text #6

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5770. Min Act: -0.1700

Data Index: 1618343 (Python Code)

Max Activating Token Index: 288

Text #7

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5770. Min Act: -0.1700

Data Index: 1572902 (Python Code)

Max Activating Token Index: 249

Text #8

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.4830. Min Act: -0.1700

Data Index: 1460871 (Python Code)

Max Activating Token Index: 538

Text #9

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.4830. Min Act: -0.1700

Data Index: 1585701 (Python Code)

Max Activating Token Index: 771

Text #10

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.4673. Min Act: -0.1700

Data Index: 1550135 (Python Code)

Max Activating Token Index: 856

Text #11

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5143. Min Act: -0.1700

Data Index: 1380704 (Python Code)

Max Activating Token Index: 213

Text #12

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5143. Min Act: -0.1700

Data Index: 1420786 (Python Code)

Max Activating Token Index: 387

Text #13

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5143. Min Act: -0.1700

Data Index: 1383219 (Python Code)

Max Activating Token Index: 566

Text #14

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5143. Min Act: -0.1700

Data Index: 1454344 (Python Code)

Max Activating Token Index: 215

Text #15

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.5143. Min Act: -0.1700

Data Index: 1419699 (Python Code)

Max Activating Token Index: 868

Text #16

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.4673. Min Act: -0.1700

Data Index: 1590332 (Python Code)

Max Activating Token Index: 708

Text #17

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.4516. Min Act: -0.1700

Data Index: 1483144 (Python Code)

Max Activating Token Index: 363

Text #18

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.4830. Min Act: -0.1696

Data Index: 1029756 (C4 (Web Text))

Max Activating Token Index: 438

Text #19

Max Range: 3.7335. Min Range: -3.7335

Max Act: 3.4830. Min Act: -0.1699

Data Index: 603243 (C4 (Web Text))

Max Activating Token Index: 879