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

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

Neuron 636 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: 5.8898. Min Range: -5.8898

Max Act: 5.8273. Min Act: -0.1700

Data Index: 515499 (C4 (Web Text))

Max Activating Token Index: 241

Text #1

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.8898. Min Act: -0.1700

Data Index: 211133 (C4 (Web Text))

Max Activating Token Index: 764

Text #2

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.8273. Min Act: -0.1700

Data Index: 410859 (C4 (Web Text))

Max Activating Token Index: 922

Text #3

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.7648. Min Act: -0.1700

Data Index: 611063 (C4 (Web Text))

Max Activating Token Index: 105

Text #4

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.8586. Min Act: -0.1700

Data Index: 993366 (C4 (Web Text))

Max Activating Token Index: 834

Text #5

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.7648. Min Act: -0.1700

Data Index: 734943 (C4 (Web Text))

Max Activating Token Index: 670

Text #6

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.8273. Min Act: -0.1700

Data Index: 1151233 (C4 (Web Text))

Max Activating Token Index: 227

Text #7

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.7648. Min Act: -0.1700

Data Index: 430957 (C4 (Web Text))

Max Activating Token Index: 649

Text #8

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.7023. Min Act: -0.1700

Data Index: 1320351 (C4 (Web Text))

Max Activating Token Index: 591

Text #9

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.7023. Min Act: -0.1700

Data Index: 1213514 (C4 (Web Text))

Max Activating Token Index: 203

Text #10

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.7023. Min Act: -0.1700

Data Index: 107426 (C4 (Web Text))

Max Activating Token Index: 726

Text #11

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.7023. Min Act: -0.1700

Data Index: 1260658 (C4 (Web Text))

Max Activating Token Index: 239

Text #12

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.6711. Min Act: -0.1700

Data Index: 514490 (C4 (Web Text))

Max Activating Token Index: 260

Text #13

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.6398. Min Act: -0.1700

Data Index: 1129519 (C4 (Web Text))

Max Activating Token Index: 196

Text #14

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.6086. Min Act: -0.1700

Data Index: 1138438 (C4 (Web Text))

Max Activating Token Index: 342

Text #15

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.6086. Min Act: -0.1700

Data Index: 785227 (C4 (Web Text))

Max Activating Token Index: 412

Text #16

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.6398. Min Act: -0.1700

Data Index: 269058 (C4 (Web Text))

Max Activating Token Index: 243

Text #17

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.5461. Min Act: -0.1700

Data Index: 602836 (C4 (Web Text))

Max Activating Token Index: 876

Text #18

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.6086. Min Act: -0.1700

Data Index: 1034932 (C4 (Web Text))

Max Activating Token Index: 858

Text #19

Max Range: 5.8898. Min Range: -5.8898

Max Act: 5.5461. Min Act: -0.1700

Data Index: 786396 (C4 (Web Text))

Max Activating Token Index: 594