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

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

Neuron 1325 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: 7.0123. Min Range: -7.0123

Max Act: 6.9811. Min Act: -0.1700

Data Index: 227419 (C4 (Web Text))

Max Activating Token Index: 904

Text #1

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.9498. Min Act: -0.1700

Data Index: 742293 (C4 (Web Text))

Max Activating Token Index: 493

Text #2

Max Range: 7.0123. Min Range: -7.0123

Max Act: 7.0123. Min Act: -0.1700

Data Index: 1103811 (C4 (Web Text))

Max Activating Token Index: 704

Text #3

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.9811. Min Act: -0.1700

Data Index: 533740 (C4 (Web Text))

Max Activating Token Index: 553

Text #4

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.9186. Min Act: -0.1700

Data Index: 1166537 (C4 (Web Text))

Max Activating Token Index: 498

Text #5

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8873. Min Act: -0.1700

Data Index: 99473 (C4 (Web Text))

Max Activating Token Index: 382

Text #6

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8873. Min Act: -0.1700

Data Index: 942360 (C4 (Web Text))

Max Activating Token Index: 49

Text #7

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.9498. Min Act: -0.1700

Data Index: 562619 (C4 (Web Text))

Max Activating Token Index: 96

Text #8

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.9186. Min Act: -0.1700

Data Index: 122181 (C4 (Web Text))

Max Activating Token Index: 528

Text #9

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.9186. Min Act: -0.1700

Data Index: 792484 (C4 (Web Text))

Max Activating Token Index: 641

Text #10

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8873. Min Act: -0.1700

Data Index: 710563 (C4 (Web Text))

Max Activating Token Index: 941

Text #11

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8873. Min Act: -0.1700

Data Index: 694536 (C4 (Web Text))

Max Activating Token Index: 695

Text #12

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8561. Min Act: -0.1700

Data Index: 191294 (C4 (Web Text))

Max Activating Token Index: 121

Text #13

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.7936. Min Act: -0.1700

Data Index: 129921 (C4 (Web Text))

Max Activating Token Index: 475

Text #14

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8873. Min Act: -0.1700

Data Index: 803617 (C4 (Web Text))

Max Activating Token Index: 1014

Text #15

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8561. Min Act: -0.1700

Data Index: 165196 (C4 (Web Text))

Max Activating Token Index: 883

Text #16

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8561. Min Act: -0.1700

Data Index: 443950 (C4 (Web Text))

Max Activating Token Index: 221

Text #17

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8248. Min Act: -0.1700

Data Index: 918162 (C4 (Web Text))

Max Activating Token Index: 169

Text #18

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.8561. Min Act: -0.1700

Data Index: 1147086 (C4 (Web Text))

Max Activating Token Index: 352

Text #19

Max Range: 7.0123. Min Range: -7.0123

Max Act: 6.7936. Min Act: -0.1700

Data Index: 753995 (C4 (Web Text))

Max Activating Token Index: 514