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

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

Neuron 1110 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: 3.8761. Min Range: -3.8761

Max Act: 3.8761. Min Act: -0.1700

Data Index: 1374355 (Python Code)

Max Activating Token Index: 291

Text #1

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.7510. Min Act: -0.1700

Data Index: 50772 (C4 (Web Text))

Max Activating Token Index: 12

Text #2

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.7510. Min Act: -0.1700

Data Index: 1584375 (Python Code)

Max Activating Token Index: 768

Text #3

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.7510. Min Act: -0.1700

Data Index: 393708 (C4 (Web Text))

Max Activating Token Index: 567

Text #4

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.7197. Min Act: -0.1700

Data Index: 694534 (C4 (Web Text))

Max Activating Token Index: 166

Text #5

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6884. Min Act: -0.1700

Data Index: 229089 (C4 (Web Text))

Max Activating Token Index: 552

Text #6

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6884. Min Act: -0.1700

Data Index: 825120 (C4 (Web Text))

Max Activating Token Index: 9

Text #7

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6884. Min Act: -0.1700

Data Index: 305147 (C4 (Web Text))

Max Activating Token Index: 681

Text #8

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6884. Min Act: -0.1700

Data Index: 174077 (C4 (Web Text))

Max Activating Token Index: 694

Text #9

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6571. Min Act: -0.1700

Data Index: 588303 (C4 (Web Text))

Max Activating Token Index: 11

Text #10

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6571. Min Act: -0.1700

Data Index: 49375 (C4 (Web Text))

Max Activating Token Index: 384

Text #11

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6571. Min Act: -0.1700

Data Index: 990549 (C4 (Web Text))

Max Activating Token Index: 972

Text #12

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6571. Min Act: -0.1700

Data Index: 85319 (C4 (Web Text))

Max Activating Token Index: 334

Text #13

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6258. Min Act: -0.1700

Data Index: 1190868 (C4 (Web Text))

Max Activating Token Index: 427

Text #14

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6258. Min Act: -0.1700

Data Index: 563747 (C4 (Web Text))

Max Activating Token Index: 797

Text #15

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.5944. Min Act: -0.1700

Data Index: 446113 (C4 (Web Text))

Max Activating Token Index: 134

Text #16

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6258. Min Act: -0.1700

Data Index: 1288262 (C4 (Web Text))

Max Activating Token Index: 109

Text #17

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.5944. Min Act: -0.1700

Data Index: 1188247 (C4 (Web Text))

Max Activating Token Index: 156

Text #18

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6258. Min Act: -0.1700

Data Index: 1068000 (C4 (Web Text))

Max Activating Token Index: 21

Text #19

Max Range: 3.8761. Min Range: -3.8761

Max Act: 3.6571. Min Act: -0.1700

Data Index: 1011907 (C4 (Web Text))

Max Activating Token Index: 645