Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1171 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.3881. Min Act: -0.1699

Data Index: 708836 (C4 (Web Text))

Max Activating Token Index: 321

Text #1

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.3725. Min Act: -0.1700

Data Index: 84150 (C4 (Web Text))

Max Activating Token Index: 11

Text #2

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.3412. Min Act: -0.1700

Data Index: 319368 (C4 (Web Text))

Max Activating Token Index: 10

Text #3

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.3412. Min Act: -0.1700

Data Index: 824255 (C4 (Web Text))

Max Activating Token Index: 27

Text #4

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.2006. Min Act: -0.1700

Data Index: 964496 (C4 (Web Text))

Max Activating Token Index: 54

Text #5

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.1537. Min Act: -0.1699

Data Index: 1309782 (C4 (Web Text))

Max Activating Token Index: 793

Text #6

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0600. Min Act: -0.1700

Data Index: 1000928 (C4 (Web Text))

Max Activating Token Index: 504

Text #7

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.1537. Min Act: -0.1700

Data Index: 657578 (C4 (Web Text))

Max Activating Token Index: 386

Text #8

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0443. Min Act: -0.1700

Data Index: 179825 (C4 (Web Text))

Max Activating Token Index: 712

Text #9

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0756. Min Act: -0.1700

Data Index: 761498 (C4 (Web Text))

Max Activating Token Index: 338

Text #10

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0912. Min Act: -0.1700

Data Index: 1117796 (C4 (Web Text))

Max Activating Token Index: 17

Text #11

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0756. Min Act: -0.1700

Data Index: 588472 (C4 (Web Text))

Max Activating Token Index: 73

Text #12

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0600. Min Act: -0.1698

Data Index: 1019528 (C4 (Web Text))

Max Activating Token Index: 382

Text #13

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0756. Min Act: -0.1700

Data Index: 47769 (C4 (Web Text))

Max Activating Token Index: 68

Text #14

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0287. Min Act: -0.1700

Data Index: 224581 (C4 (Web Text))

Max Activating Token Index: 455

Text #15

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.1068. Min Act: -0.1700

Data Index: 853749 (C4 (Web Text))

Max Activating Token Index: 372

Text #16

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0287. Min Act: -0.1700

Data Index: 1170861 (C4 (Web Text))

Max Activating Token Index: 505

Text #17

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0443. Min Act: -0.1700

Data Index: 1135412 (C4 (Web Text))

Max Activating Token Index: 69

Text #18

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0443. Min Act: -0.1700

Data Index: 212509 (C4 (Web Text))

Max Activating Token Index: 75

Text #19

Max Range: 5.3881. Min Range: -5.3881

Max Act: 5.0131. Min Act: -0.1700

Data Index: 1456487 (Python Code)

Max Activating Token Index: 860