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

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

Neuron 1224 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: 5.1459. Min Range: -5.1459

Max Act: 5.0209. Min Act: -0.1700

Data Index: 1169029 (C4 (Web Text))

Max Activating Token Index: 719

Text #1

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.9897. Min Act: -0.1700

Data Index: 1495125 (Python Code)

Max Activating Token Index: 492

Text #2

Max Range: 5.1459. Min Range: -5.1459

Max Act: 5.1459. Min Act: -0.1700

Data Index: 1023585 (C4 (Web Text))

Max Activating Token Index: 715

Text #3

Max Range: 5.1459. Min Range: -5.1459

Max Act: 5.0522. Min Act: -0.1700

Data Index: 1006749 (C4 (Web Text))

Max Activating Token Index: 747

Text #4

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8022. Min Act: -0.1700

Data Index: 540951 (C4 (Web Text))

Max Activating Token Index: 628

Text #5

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8647. Min Act: -0.1700

Data Index: 773782 (C4 (Web Text))

Max Activating Token Index: 519

Text #6

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8959. Min Act: -0.1700

Data Index: 205661 (C4 (Web Text))

Max Activating Token Index: 196

Text #7

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8022. Min Act: -0.1700

Data Index: 750799 (C4 (Web Text))

Max Activating Token Index: 382

Text #8

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.7709. Min Act: -0.1700

Data Index: 769119 (C4 (Web Text))

Max Activating Token Index: 619

Text #9

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8334. Min Act: -0.1700

Data Index: 1591367 (Python Code)

Max Activating Token Index: 202

Text #10

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8647. Min Act: -0.1700

Data Index: 484089 (C4 (Web Text))

Max Activating Token Index: 308

Text #11

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8022. Min Act: -0.1700

Data Index: 1258816 (C4 (Web Text))

Max Activating Token Index: 567

Text #12

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8647. Min Act: -0.1700

Data Index: 67265 (C4 (Web Text))

Max Activating Token Index: 603

Text #13

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.7397. Min Act: -0.1700

Data Index: 1185597 (C4 (Web Text))

Max Activating Token Index: 142

Text #14

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8334. Min Act: -0.1700

Data Index: 924100 (C4 (Web Text))

Max Activating Token Index: 680

Text #15

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8647. Min Act: -0.1700

Data Index: 884131 (C4 (Web Text))

Max Activating Token Index: 366

Text #16

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.7397. Min Act: -0.1700

Data Index: 667472 (C4 (Web Text))

Max Activating Token Index: 629

Text #17

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8647. Min Act: -0.1700

Data Index: 140063 (C4 (Web Text))

Max Activating Token Index: 713

Text #18

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.8022. Min Act: -0.1700

Data Index: 596953 (C4 (Web Text))

Max Activating Token Index: 118

Text #19

Max Range: 5.1459. Min Range: -5.1459

Max Act: 4.7397. Min Act: -0.1700

Data Index: 840519 (C4 (Web Text))

Max Activating Token Index: 862