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

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

Neuron 573 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.7073. Min Act: -0.1700

Data Index: 518541 (C4 (Web Text))

Max Activating Token Index: 64

Text #1

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.6117. Min Act: -0.1700

Data Index: 1011725 (C4 (Web Text))

Max Activating Token Index: 148

Text #2

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5855. Min Act: -0.1700

Data Index: 827990 (C4 (Web Text))

Max Activating Token Index: 208

Text #3

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5768. Min Act: -0.1700

Data Index: 103216 (C4 (Web Text))

Max Activating Token Index: 193

Text #4

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5330. Min Act: -0.1700

Data Index: 1655122 (Python Code)

Max Activating Token Index: 127

Text #5

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5418. Min Act: -0.1700

Data Index: 756590 (C4 (Web Text))

Max Activating Token Index: 156

Text #6

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5418. Min Act: -0.1700

Data Index: 1158022 (C4 (Web Text))

Max Activating Token Index: 100

Text #7

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5243. Min Act: -0.1700

Data Index: 893644 (C4 (Web Text))

Max Activating Token Index: 103

Text #8

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5243. Min Act: -0.1700

Data Index: 76820 (C4 (Web Text))

Max Activating Token Index: 152

Text #9

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5155. Min Act: -0.1700

Data Index: 124097 (C4 (Web Text))

Max Activating Token Index: 150

Text #10

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5155. Min Act: -0.1700

Data Index: 580453 (C4 (Web Text))

Max Activating Token Index: 159

Text #11

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5155. Min Act: -0.1700

Data Index: 616279 (C4 (Web Text))

Max Activating Token Index: 149

Text #12

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5155. Min Act: -0.1700

Data Index: 566274 (C4 (Web Text))

Max Activating Token Index: 148

Text #13

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5155. Min Act: -0.1700

Data Index: 840285 (C4 (Web Text))

Max Activating Token Index: 148

Text #14

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5067. Min Act: -0.1700

Data Index: 951815 (C4 (Web Text))

Max Activating Token Index: 127

Text #15

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5067. Min Act: -0.1699

Data Index: 646040 (C4 (Web Text))

Max Activating Token Index: 138

Text #16

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5067. Min Act: -0.1699

Data Index: 425428 (C4 (Web Text))

Max Activating Token Index: 95

Text #17

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.4980. Min Act: -0.1700

Data Index: 1017670 (C4 (Web Text))

Max Activating Token Index: 116

Text #18

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.4980. Min Act: -0.1700

Data Index: 220159 (C4 (Web Text))

Max Activating Token Index: 176

Text #19

Max Range: 1.7073. Min Range: -1.7073

Max Act: 1.5067. Min Act: -0.1700

Data Index: 215504 (C4 (Web Text))

Max Activating Token Index: 149