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

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

Neuron 1972 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: 5.2338. Min Range: -5.2338

Max Act: 5.2338. Min Act: -0.1700

Data Index: 671386 (C4 (Web Text))

Max Activating Token Index: 56

Text #1

Max Range: 5.2338. Min Range: -5.2338

Max Act: 5.1401. Min Act: -0.1700

Data Index: 470440 (C4 (Web Text))

Max Activating Token Index: 542

Text #2

Max Range: 5.2338. Min Range: -5.2338

Max Act: 5.2026. Min Act: -0.1700

Data Index: 516041 (C4 (Web Text))

Max Activating Token Index: 845

Text #3

Max Range: 5.2338. Min Range: -5.2338

Max Act: 5.1401. Min Act: -0.1700

Data Index: 596850 (C4 (Web Text))

Max Activating Token Index: 882

Text #4

Max Range: 5.2338. Min Range: -5.2338

Max Act: 5.1401. Min Act: -0.1699

Data Index: 216495 (C4 (Web Text))

Max Activating Token Index: 566

Text #5

Max Range: 5.2338. Min Range: -5.2338

Max Act: 5.0776. Min Act: -0.1700

Data Index: 1288483 (C4 (Web Text))

Max Activating Token Index: 441

Text #6

Max Range: 5.2338. Min Range: -5.2338

Max Act: 5.1088. Min Act: -0.1700

Data Index: 1268740 (C4 (Web Text))

Max Activating Token Index: 787

Text #7

Max Range: 5.2338. Min Range: -5.2338

Max Act: 5.0151. Min Act: -0.1700

Data Index: 400636 (C4 (Web Text))

Max Activating Token Index: 830

Text #8

Max Range: 5.2338. Min Range: -5.2338

Max Act: 5.0151. Min Act: -0.1700

Data Index: 378003 (C4 (Web Text))

Max Activating Token Index: 768

Text #9

Max Range: 5.2338. Min Range: -5.2338

Max Act: 5.0151. Min Act: -0.1700

Data Index: 1125855 (C4 (Web Text))

Max Activating Token Index: 806

Text #10

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.9526. Min Act: -0.1700

Data Index: 97777 (C4 (Web Text))

Max Activating Token Index: 362

Text #11

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.9838. Min Act: -0.1700

Data Index: 1025305 (C4 (Web Text))

Max Activating Token Index: 230

Text #12

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.9526. Min Act: -0.1700

Data Index: 700970 (C4 (Web Text))

Max Activating Token Index: 745

Text #13

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.9838. Min Act: -0.1700

Data Index: 109933 (C4 (Web Text))

Max Activating Token Index: 382

Text #14

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.9213. Min Act: -0.1700

Data Index: 1283311 (C4 (Web Text))

Max Activating Token Index: 539

Text #15

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.9213. Min Act: -0.1700

Data Index: 1334433 (C4 (Web Text))

Max Activating Token Index: 581

Text #16

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.9213. Min Act: -0.1700

Data Index: 325654 (C4 (Web Text))

Max Activating Token Index: 778

Text #17

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.9213. Min Act: -0.1700

Data Index: 470267 (C4 (Web Text))

Max Activating Token Index: 126

Text #18

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.9213. Min Act: -0.1700

Data Index: 592684 (C4 (Web Text))

Max Activating Token Index: 125

Text #19

Max Range: 5.2338. Min Range: -5.2338

Max Act: 4.8901. Min Act: -0.1700

Data Index: 335403 (C4 (Web Text))

Max Activating Token Index: 680