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

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

Neuron 1138 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: 8.3860. Min Range: -8.3860

Max Act: 8.3860. Min Act: -0.1700

Data Index: 930295 (C4 (Web Text))

Max Activating Token Index: 658

Text #1

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.3235. Min Act: -0.1700

Data Index: 82983 (C4 (Web Text))

Max Activating Token Index: 487

Text #2

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.3860. Min Act: -0.1700

Data Index: 1128186 (C4 (Web Text))

Max Activating Token Index: 528

Text #3

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.2610. Min Act: -0.1700

Data Index: 534332 (C4 (Web Text))

Max Activating Token Index: 237

Text #4

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.2610. Min Act: -0.1700

Data Index: 42762 (C4 (Web Text))

Max Activating Token Index: 480

Text #5

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1985. Min Act: -0.1700

Data Index: 1021717 (C4 (Web Text))

Max Activating Token Index: 487

Text #6

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1985. Min Act: -0.1700

Data Index: 556610 (C4 (Web Text))

Max Activating Token Index: 252

Text #7

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1360. Min Act: -0.1700

Data Index: 336249 (C4 (Web Text))

Max Activating Token Index: 246

Text #8

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1360. Min Act: -0.1700

Data Index: 1250803 (C4 (Web Text))

Max Activating Token Index: 764

Text #9

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.0110. Min Act: -0.1700

Data Index: 873019 (C4 (Web Text))

Max Activating Token Index: 821

Text #10

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1985. Min Act: -0.1700

Data Index: 466553 (C4 (Web Text))

Max Activating Token Index: 557

Text #11

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.0735. Min Act: -0.1700

Data Index: 1036109 (C4 (Web Text))

Max Activating Token Index: 558

Text #12

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.0735. Min Act: -0.1700

Data Index: 479687 (C4 (Web Text))

Max Activating Token Index: 654

Text #13

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1360. Min Act: -0.1700

Data Index: 134784 (C4 (Web Text))

Max Activating Token Index: 214

Text #14

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1985. Min Act: -0.1700

Data Index: 392358 (C4 (Web Text))

Max Activating Token Index: 487

Text #15

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1985. Min Act: -0.1700

Data Index: 519323 (C4 (Web Text))

Max Activating Token Index: 338

Text #16

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.0110. Min Act: -0.1700

Data Index: 14140 (C4 (Web Text))

Max Activating Token Index: 738

Text #17

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1985. Min Act: -0.1700

Data Index: 965018 (C4 (Web Text))

Max Activating Token Index: 283

Text #18

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.1985. Min Act: -0.1700

Data Index: 1147226 (C4 (Web Text))

Max Activating Token Index: 706

Text #19

Max Range: 8.3860. Min Range: -8.3860

Max Act: 8.0110. Min Act: -0.1700

Data Index: 2524 (C4 (Web Text))

Max Activating Token Index: 648