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

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

Neuron 1306 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.4233. Min Act: -0.1700

Data Index: 760917 (C4 (Web Text))

Max Activating Token Index: 794

Text #1

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.4233. Min Act: -0.1700

Data Index: 49490 (C4 (Web Text))

Max Activating Token Index: 835

Text #2

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.3608. Min Act: -0.1700

Data Index: 1194174 (C4 (Web Text))

Max Activating Token Index: 525

Text #3

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2983. Min Act: -0.1700

Data Index: 64791 (C4 (Web Text))

Max Activating Token Index: 512

Text #4

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2670. Min Act: -0.1700

Data Index: 1091694 (C4 (Web Text))

Max Activating Token Index: 477

Text #5

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2670. Min Act: -0.1700

Data Index: 789621 (C4 (Web Text))

Max Activating Token Index: 688

Text #6

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2358. Min Act: -0.1700

Data Index: 813573 (C4 (Web Text))

Max Activating Token Index: 356

Text #7

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2045. Min Act: -0.1700

Data Index: 1076961 (C4 (Web Text))

Max Activating Token Index: 686

Text #8

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2045. Min Act: -0.1700

Data Index: 951930 (C4 (Web Text))

Max Activating Token Index: 510

Text #9

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2045. Min Act: -0.1700

Data Index: 952432 (C4 (Web Text))

Max Activating Token Index: 641

Text #10

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2045. Min Act: -0.1700

Data Index: 288083 (C4 (Web Text))

Max Activating Token Index: 258

Text #11

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2358. Min Act: -0.1700

Data Index: 53178 (C4 (Web Text))

Max Activating Token Index: 608

Text #12

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.2045. Min Act: -0.1700

Data Index: 553699 (C4 (Web Text))

Max Activating Token Index: 538

Text #13

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.1733. Min Act: -0.1700

Data Index: 238306 (C4 (Web Text))

Max Activating Token Index: 497

Text #14

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.1733. Min Act: -0.1700

Data Index: 293010 (C4 (Web Text))

Max Activating Token Index: 750

Text #15

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.1733. Min Act: -0.1700

Data Index: 1143767 (C4 (Web Text))

Max Activating Token Index: 364

Text #16

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.1733. Min Act: -0.1700

Data Index: 1269608 (C4 (Web Text))

Max Activating Token Index: 461

Text #17

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.1733. Min Act: -0.1700

Data Index: 421203 (C4 (Web Text))

Max Activating Token Index: 562

Text #18

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.1108. Min Act: -0.1700

Data Index: 622560 (C4 (Web Text))

Max Activating Token Index: 594

Text #19

Max Range: 5.4233. Min Range: -5.4233

Max Act: 5.1420. Min Act: -0.1700

Data Index: 92553 (C4 (Web Text))

Max Activating Token Index: 975