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

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

Neuron 91 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.5349. Min Act: -0.1700

Data Index: 440419 (C4 (Web Text))

Max Activating Token Index: 508

Text #1

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.5035. Min Act: -0.1700

Data Index: 420676 (C4 (Web Text))

Max Activating Token Index: 950

Text #2

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4879. Min Act: -0.1700

Data Index: 381852 (C4 (Web Text))

Max Activating Token Index: 529

Text #3

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.5035. Min Act: -0.1700

Data Index: 1311012 (C4 (Web Text))

Max Activating Token Index: 909

Text #4

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.5035. Min Act: -0.1700

Data Index: 876628 (C4 (Web Text))

Max Activating Token Index: 266

Text #5

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4722. Min Act: -0.1700

Data Index: 1323026 (C4 (Web Text))

Max Activating Token Index: 232

Text #6

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4565. Min Act: -0.1700

Data Index: 522791 (C4 (Web Text))

Max Activating Token Index: 297

Text #7

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4565. Min Act: -0.1700

Data Index: 527714 (C4 (Web Text))

Max Activating Token Index: 562

Text #8

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4408. Min Act: -0.1700

Data Index: 943521 (C4 (Web Text))

Max Activating Token Index: 575

Text #9

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4565. Min Act: -0.1700

Data Index: 492212 (C4 (Web Text))

Max Activating Token Index: 595

Text #10

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4565. Min Act: -0.1700

Data Index: 461796 (C4 (Web Text))

Max Activating Token Index: 688

Text #11

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4408. Min Act: -0.1700

Data Index: 435895 (C4 (Web Text))

Max Activating Token Index: 631

Text #12

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4095. Min Act: -0.1700

Data Index: 477528 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4252. Min Act: -0.1700

Data Index: 1390242 (Python Code)

Max Activating Token Index: 416

Text #14

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.3938. Min Act: -0.1700

Data Index: 577870 (C4 (Web Text))

Max Activating Token Index: 617

Text #15

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4095. Min Act: -0.1700

Data Index: 136798 (C4 (Web Text))

Max Activating Token Index: 804

Text #16

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4095. Min Act: -0.1700

Data Index: 15515 (C4 (Web Text))

Max Activating Token Index: 231

Text #17

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4095. Min Act: -0.1700

Data Index: 334645 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.4408. Min Act: -0.1700

Data Index: 774633 (C4 (Web Text))

Max Activating Token Index: 69

Text #19

Max Range: 3.5349. Min Range: -3.5349

Max Act: 3.3938. Min Act: -0.1700

Data Index: 1175558 (C4 (Web Text))

Max Activating Token Index: 991