Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1508 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 7.0211. Min Range: -7.0211

Max Act: 7.0211. Min Act: -0.1700

Data Index: 614341 (C4 (Web Text))

Max Activating Token Index: 428

Text #1

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.9586. Min Act: -0.1700

Data Index: 437995 (C4 (Web Text))

Max Activating Token Index: 749

Text #2

Max Range: 7.0211. Min Range: -7.0211

Max Act: 7.0211. Min Act: -0.1700

Data Index: 347207 (C4 (Web Text))

Max Activating Token Index: 659

Text #3

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.9273. Min Act: -0.1700

Data Index: 446291 (C4 (Web Text))

Max Activating Token Index: 309

Text #4

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.8961. Min Act: -0.1700

Data Index: 1302641 (C4 (Web Text))

Max Activating Token Index: 995

Text #5

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.7711. Min Act: -0.1700

Data Index: 78025 (C4 (Web Text))

Max Activating Token Index: 1006

Text #6

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.6773. Min Act: -0.1700

Data Index: 82986 (C4 (Web Text))

Max Activating Token Index: 340

Text #7

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.6148. Min Act: -0.1700

Data Index: 480158 (C4 (Web Text))

Max Activating Token Index: 120

Text #8

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.5836. Min Act: -0.1700

Data Index: 841093 (C4 (Web Text))

Max Activating Token Index: 614

Text #9

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.5836. Min Act: -0.1700

Data Index: 1463516 (Python Code)

Max Activating Token Index: 330

Text #10

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.5836. Min Act: -0.1700

Data Index: 642768 (C4 (Web Text))

Max Activating Token Index: 204

Text #11

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.5211. Min Act: -0.1700

Data Index: 963900 (C4 (Web Text))

Max Activating Token Index: 357

Text #12

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.5523. Min Act: -0.1700

Data Index: 831236 (C4 (Web Text))

Max Activating Token Index: 1022

Text #13

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.5836. Min Act: -0.1700

Data Index: 1155686 (C4 (Web Text))

Max Activating Token Index: 757

Text #14

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.5523. Min Act: -0.1700

Data Index: 1251325 (C4 (Web Text))

Max Activating Token Index: 367

Text #15

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.5523. Min Act: -0.1700

Data Index: 949213 (C4 (Web Text))

Max Activating Token Index: 446

Text #16

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.5211. Min Act: -0.1700

Data Index: 229707 (C4 (Web Text))

Max Activating Token Index: 329

Text #17

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.4586. Min Act: -0.1700

Data Index: 769640 (C4 (Web Text))

Max Activating Token Index: 320

Text #18

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.4898. Min Act: -0.1700

Data Index: 490418 (C4 (Web Text))

Max Activating Token Index: 647

Text #19

Max Range: 7.0211. Min Range: -7.0211

Max Act: 6.4586. Min Act: -0.1700

Data Index: 1621372 (Python Code)

Max Activating Token Index: 1019