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

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

Neuron 1423 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.0161. Min Range: -7.0161

Max Act: 7.0161. Min Act: -0.1700

Data Index: 368375 (C4 (Web Text))

Max Activating Token Index: 90

Text #1

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.8286. Min Act: -0.1700

Data Index: 1255383 (C4 (Web Text))

Max Activating Token Index: 76

Text #2

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.8286. Min Act: -0.1700

Data Index: 591576 (C4 (Web Text))

Max Activating Token Index: 99

Text #3

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.6724. Min Act: -0.1700

Data Index: 1024314 (C4 (Web Text))

Max Activating Token Index: 150

Text #4

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.6724. Min Act: -0.1700

Data Index: 683297 (C4 (Web Text))

Max Activating Token Index: 1016

Text #5

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.6411. Min Act: -0.1700

Data Index: 1048227 (C4 (Web Text))

Max Activating Token Index: 104

Text #6

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.5161. Min Act: -0.1700

Data Index: 131676 (C4 (Web Text))

Max Activating Token Index: 212

Text #7

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.5786. Min Act: -0.1700

Data Index: 1052186 (C4 (Web Text))

Max Activating Token Index: 172

Text #8

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.5474. Min Act: -0.1700

Data Index: 362950 (C4 (Web Text))

Max Activating Token Index: 102

Text #9

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.4849. Min Act: -0.1700

Data Index: 459691 (C4 (Web Text))

Max Activating Token Index: 1004

Text #10

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.5161. Min Act: -0.1700

Data Index: 72224 (C4 (Web Text))

Max Activating Token Index: 1016

Text #11

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.4849. Min Act: -0.1700

Data Index: 1117471 (C4 (Web Text))

Max Activating Token Index: 245

Text #12

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.4849. Min Act: -0.1700

Data Index: 394849 (C4 (Web Text))

Max Activating Token Index: 301

Text #13

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.4536. Min Act: -0.1700

Data Index: 998424 (C4 (Web Text))

Max Activating Token Index: 187

Text #14

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.4224. Min Act: -0.1700

Data Index: 567349 (C4 (Web Text))

Max Activating Token Index: 1018

Text #15

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.3599. Min Act: -0.1700

Data Index: 1112183 (C4 (Web Text))

Max Activating Token Index: 845

Text #16

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.3911. Min Act: -0.1700

Data Index: 641181 (C4 (Web Text))

Max Activating Token Index: 299

Text #17

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.3599. Min Act: -0.1700

Data Index: 150673 (C4 (Web Text))

Max Activating Token Index: 342

Text #18

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.3286. Min Act: -0.1700

Data Index: 1033729 (C4 (Web Text))

Max Activating Token Index: 940

Text #19

Max Range: 7.0161. Min Range: -7.0161

Max Act: 6.4224. Min Act: -0.1700

Data Index: 1090618 (C4 (Web Text))

Max Activating Token Index: 753