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

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

Neuron 973 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: 3.5430. Min Range: -3.5430

Max Act: 3.5430. Min Act: -0.1700

Data Index: 742989 (C4 (Web Text))

Max Activating Token Index: 924

Text #1

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.5116. Min Act: -0.1700

Data Index: 1058126 (C4 (Web Text))

Max Activating Token Index: 743

Text #2

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.4176. Min Act: -0.1700

Data Index: 1448140 (Python Code)

Max Activating Token Index: 816

Text #3

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.3862. Min Act: -0.1700

Data Index: 1396954 (Python Code)

Max Activating Token Index: 407

Text #4

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.3234. Min Act: -0.1700

Data Index: 1421861 (Python Code)

Max Activating Token Index: 853

Text #5

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.3234. Min Act: -0.1700

Data Index: 623036 (C4 (Web Text))

Max Activating Token Index: 372

Text #6

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.3548. Min Act: -0.1700

Data Index: 557507 (C4 (Web Text))

Max Activating Token Index: 507

Text #7

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.2606. Min Act: -0.1700

Data Index: 747334 (C4 (Web Text))

Max Activating Token Index: 863

Text #8

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.2606. Min Act: -0.1700

Data Index: 1518451 (Python Code)

Max Activating Token Index: 226

Text #9

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.2606. Min Act: -0.1700

Data Index: 1654371 (Python Code)

Max Activating Token Index: 387

Text #10

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.2920. Min Act: -0.1700

Data Index: 1542323 (Python Code)

Max Activating Token Index: 294

Text #11

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.2920. Min Act: -0.1700

Data Index: 1599306 (Python Code)

Max Activating Token Index: 277

Text #12

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.3548. Min Act: -0.1700

Data Index: 193517 (C4 (Web Text))

Max Activating Token Index: 850

Text #13

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.3548. Min Act: -0.1700

Data Index: 883050 (C4 (Web Text))

Max Activating Token Index: 288

Text #14

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.3234. Min Act: -0.1700

Data Index: 893256 (C4 (Web Text))

Max Activating Token Index: 601

Text #15

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.2920. Min Act: -0.1700

Data Index: 1438759 (Python Code)

Max Activating Token Index: 476

Text #16

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.3234. Min Act: -0.1700

Data Index: 1537163 (Python Code)

Max Activating Token Index: 607

Text #17

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.2606. Min Act: -0.1700

Data Index: 805667 (C4 (Web Text))

Max Activating Token Index: 149

Text #18

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.2920. Min Act: -0.1700

Data Index: 782158 (C4 (Web Text))

Max Activating Token Index: 965

Text #19

Max Range: 3.5430. Min Range: -3.5430

Max Act: 3.2920. Min Act: -0.1700

Data Index: 747310 (C4 (Web Text))

Max Activating Token Index: 556