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

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

Neuron 1097 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: 8.8324. Min Range: -8.8324

Max Act: 8.8324. Min Act: -0.1700

Data Index: 1311774 (C4 (Web Text))

Max Activating Token Index: 493

Text #1

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.7074. Min Act: -0.1700

Data Index: 175952 (C4 (Web Text))

Max Activating Token Index: 311

Text #2

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.7699. Min Act: -0.1700

Data Index: 481721 (C4 (Web Text))

Max Activating Token Index: 703

Text #3

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.4574. Min Act: -0.1700

Data Index: 1304129 (C4 (Web Text))

Max Activating Token Index: 210

Text #4

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.3949. Min Act: -0.1700

Data Index: 564066 (C4 (Web Text))

Max Activating Token Index: 502

Text #5

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.3949. Min Act: -0.1699

Data Index: 502389 (C4 (Web Text))

Max Activating Token Index: 538

Text #6

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.3949. Min Act: -0.1700

Data Index: 1076371 (C4 (Web Text))

Max Activating Token Index: 258

Text #7

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.3949. Min Act: -0.1700

Data Index: 1179127 (C4 (Web Text))

Max Activating Token Index: 195

Text #8

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.3949. Min Act: -0.1699

Data Index: 1086805 (C4 (Web Text))

Max Activating Token Index: 890

Text #9

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.1449. Min Act: -0.1700

Data Index: 783890 (C4 (Web Text))

Max Activating Token Index: 116

Text #10

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.1449. Min Act: -0.1699

Data Index: 1016167 (C4 (Web Text))

Max Activating Token Index: 813

Text #11

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.0199. Min Act: -0.1700

Data Index: 1208400 (C4 (Web Text))

Max Activating Token Index: 513

Text #12

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.0824. Min Act: -0.1693

Data Index: 588656 (C4 (Web Text))

Max Activating Token Index: 130

Text #13

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.1449. Min Act: -0.1700

Data Index: 657185 (C4 (Web Text))

Max Activating Token Index: 393

Text #14

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.0824. Min Act: -0.1700

Data Index: 1036657 (C4 (Web Text))

Max Activating Token Index: 537

Text #15

Max Range: 8.8324. Min Range: -8.8324

Max Act: 8.0199. Min Act: -0.1700

Data Index: 861051 (C4 (Web Text))

Max Activating Token Index: 206

Text #16

Max Range: 8.8324. Min Range: -8.8324

Max Act: 7.9574. Min Act: -0.1700

Data Index: 209667 (C4 (Web Text))

Max Activating Token Index: 31

Text #17

Max Range: 8.8324. Min Range: -8.8324

Max Act: 7.9574. Min Act: -0.1700

Data Index: 1207813 (C4 (Web Text))

Max Activating Token Index: 220

Text #18

Max Range: 8.8324. Min Range: -8.8324

Max Act: 7.8949. Min Act: -0.1700

Data Index: 627040 (C4 (Web Text))

Max Activating Token Index: 301

Text #19

Max Range: 8.8324. Min Range: -8.8324

Max Act: 7.9574. Min Act: -0.1700

Data Index: 717396 (C4 (Web Text))

Max Activating Token Index: 540