Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 949 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.1706. Min Range: -5.1706

Max Act: 5.1706. Min Act: -0.1700

Data Index: 478756 (C4 (Web Text))

Max Activating Token Index: 30

Text #1

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.9988. Min Act: -0.1700

Data Index: 1231887 (C4 (Web Text))

Max Activating Token Index: 13

Text #2

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.9675. Min Act: -0.1700

Data Index: 416290 (C4 (Web Text))

Max Activating Token Index: 886

Text #3

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.9050. Min Act: -0.1700

Data Index: 940454 (C4 (Web Text))

Max Activating Token Index: 837

Text #4

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8425. Min Act: -0.1700

Data Index: 1019309 (C4 (Web Text))

Max Activating Token Index: 230

Text #5

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8581. Min Act: -0.1700

Data Index: 1087186 (C4 (Web Text))

Max Activating Token Index: 322

Text #6

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8425. Min Act: -0.1699

Data Index: 486263 (C4 (Web Text))

Max Activating Token Index: 891

Text #7

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8425. Min Act: -0.1700

Data Index: 1473924 (Python Code)

Max Activating Token Index: 379

Text #8

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8269. Min Act: -0.1700

Data Index: 1433610 (Python Code)

Max Activating Token Index: 741

Text #9

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8113. Min Act: -0.1700

Data Index: 1446801 (Python Code)

Max Activating Token Index: 266

Text #10

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8269. Min Act: -0.1700

Data Index: 1553137 (Python Code)

Max Activating Token Index: 145

Text #11

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8113. Min Act: -0.1700

Data Index: 979959 (C4 (Web Text))

Max Activating Token Index: 875

Text #12

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.7956. Min Act: -0.1700

Data Index: 1423247 (Python Code)

Max Activating Token Index: 23

Text #13

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.7800. Min Act: -0.1700

Data Index: 1408174 (Python Code)

Max Activating Token Index: 556

Text #14

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8113. Min Act: -0.1700

Data Index: 325587 (C4 (Web Text))

Max Activating Token Index: 248

Text #15

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.8113. Min Act: -0.1700

Data Index: 271453 (C4 (Web Text))

Max Activating Token Index: 291

Text #16

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.7800. Min Act: -0.1700

Data Index: 1576543 (Python Code)

Max Activating Token Index: 765

Text #17

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.7800. Min Act: -0.1700

Data Index: 624691 (C4 (Web Text))

Max Activating Token Index: 103

Text #18

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.7800. Min Act: -0.1700

Data Index: 329866 (C4 (Web Text))

Max Activating Token Index: 84

Text #19

Max Range: 5.1706. Min Range: -5.1706

Max Act: 4.7488. Min Act: -0.1700

Data Index: 596167 (C4 (Web Text))

Max Activating Token Index: 562