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

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

Neuron 665 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: 5.0295. Min Range: -5.0295

Max Act: 5.0295. Min Act: -0.1700

Data Index: 1029255 (C4 (Web Text))

Max Activating Token Index: 167

Text #1

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.9670. Min Act: -0.1700

Data Index: 858801 (C4 (Web Text))

Max Activating Token Index: 161

Text #2

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.8733. Min Act: -0.1700

Data Index: 977357 (C4 (Web Text))

Max Activating Token Index: 978

Text #3

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.8733. Min Act: -0.1700

Data Index: 54125 (C4 (Web Text))

Max Activating Token Index: 152

Text #4

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.8108. Min Act: -0.1700

Data Index: 1200958 (C4 (Web Text))

Max Activating Token Index: 185

Text #5

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.7795. Min Act: -0.1700

Data Index: 612658 (C4 (Web Text))

Max Activating Token Index: 382

Text #6

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.7483. Min Act: -0.1700

Data Index: 139059 (C4 (Web Text))

Max Activating Token Index: 155

Text #7

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.7795. Min Act: -0.1700

Data Index: 362132 (C4 (Web Text))

Max Activating Token Index: 376

Text #8

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.7483. Min Act: -0.1700

Data Index: 511348 (C4 (Web Text))

Max Activating Token Index: 46

Text #9

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.7795. Min Act: -0.1700

Data Index: 95587 (C4 (Web Text))

Max Activating Token Index: 40

Text #10

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.7483. Min Act: -0.1700

Data Index: 1028632 (C4 (Web Text))

Max Activating Token Index: 11

Text #11

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.7170. Min Act: -0.1695

Data Index: 889628 (C4 (Web Text))

Max Activating Token Index: 87

Text #12

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.6858. Min Act: -0.1700

Data Index: 446221 (C4 (Web Text))

Max Activating Token Index: 57

Text #13

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.6545. Min Act: -0.1697

Data Index: 987069 (C4 (Web Text))

Max Activating Token Index: 173

Text #14

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.6545. Min Act: -0.1700

Data Index: 42040 (C4 (Web Text))

Max Activating Token Index: 746

Text #15

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.6545. Min Act: -0.1700

Data Index: 194588 (C4 (Web Text))

Max Activating Token Index: 570

Text #16

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.6389. Min Act: -0.1700

Data Index: 773482 (C4 (Web Text))

Max Activating Token Index: 541

Text #17

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.6545. Min Act: -0.1700

Data Index: 1299315 (C4 (Web Text))

Max Activating Token Index: 647

Text #18

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.6545. Min Act: -0.1700

Data Index: 275104 (C4 (Web Text))

Max Activating Token Index: 190

Text #19

Max Range: 5.0295. Min Range: -5.0295

Max Act: 4.6389. Min Act: -0.1700

Data Index: 836228 (C4 (Web Text))

Max Activating Token Index: 674