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

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

Neuron 1750 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.2003. Min Act: -0.1700

Data Index: 346731 (C4 (Web Text))

Max Activating Token Index: 700

Text #1

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.2003. Min Act: -0.1700

Data Index: 800093 (C4 (Web Text))

Max Activating Token Index: 589

Text #2

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.1688. Min Act: -0.1700

Data Index: 223726 (C4 (Web Text))

Max Activating Token Index: 215

Text #3

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0900. Min Act: -0.1700

Data Index: 630685 (C4 (Web Text))

Max Activating Token Index: 374

Text #4

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.1531. Min Act: -0.1700

Data Index: 590388 (C4 (Web Text))

Max Activating Token Index: 362

Text #5

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.1058. Min Act: -0.1700

Data Index: 13481 (C4 (Web Text))

Max Activating Token Index: 924

Text #6

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.1058. Min Act: -0.1700

Data Index: 588950 (C4 (Web Text))

Max Activating Token Index: 644

Text #7

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0743. Min Act: -0.1700

Data Index: 584606 (C4 (Web Text))

Max Activating Token Index: 353

Text #8

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0585. Min Act: -0.1700

Data Index: 672316 (C4 (Web Text))

Max Activating Token Index: 154

Text #9

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0900. Min Act: -0.1700

Data Index: 1126511 (C4 (Web Text))

Max Activating Token Index: 642

Text #10

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0900. Min Act: -0.1700

Data Index: 1295178 (C4 (Web Text))

Max Activating Token Index: 496

Text #11

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0900. Min Act: -0.1700

Data Index: 257885 (C4 (Web Text))

Max Activating Token Index: 950

Text #12

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0269. Min Act: -0.1700

Data Index: 1126991 (C4 (Web Text))

Max Activating Token Index: 674

Text #13

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0585. Min Act: -0.1700

Data Index: 1263951 (C4 (Web Text))

Max Activating Token Index: 150

Text #14

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0585. Min Act: -0.1700

Data Index: 391670 (C4 (Web Text))

Max Activating Token Index: 465

Text #15

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0427. Min Act: -0.1700

Data Index: 157686 (C4 (Web Text))

Max Activating Token Index: 669

Text #16

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0427. Min Act: -0.1700

Data Index: 654744 (C4 (Web Text))

Max Activating Token Index: 566

Text #17

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0743. Min Act: -0.1700

Data Index: 92328 (C4 (Web Text))

Max Activating Token Index: 599

Text #18

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0269. Min Act: -0.1700

Data Index: 74009 (C4 (Web Text))

Max Activating Token Index: 676

Text #19

Max Range: 3.2003. Min Range: -3.2003

Max Act: 3.0427. Min Act: -0.1700

Data Index: 99367 (C4 (Web Text))

Max Activating Token Index: 115