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

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

Neuron 1529 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: 5.4524. Min Range: -5.4524

Max Act: 5.4524. Min Act: -0.1700

Data Index: 1274273 (C4 (Web Text))

Max Activating Token Index: 180

Text #1

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.4524. Min Act: -0.1700

Data Index: 777325 (C4 (Web Text))

Max Activating Token Index: 405

Text #2

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.3274. Min Act: -0.1700

Data Index: 1215675 (C4 (Web Text))

Max Activating Token Index: 769

Text #3

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2962. Min Act: -0.1700

Data Index: 1350173 (C4 (Web Text))

Max Activating Token Index: 205

Text #4

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.3274. Min Act: -0.1700

Data Index: 456167 (C4 (Web Text))

Max Activating Token Index: 859

Text #5

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.3274. Min Act: -0.1700

Data Index: 573352 (C4 (Web Text))

Max Activating Token Index: 420

Text #6

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2962. Min Act: -0.1699

Data Index: 562987 (C4 (Web Text))

Max Activating Token Index: 988

Text #7

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2337. Min Act: -0.1700

Data Index: 17206 (C4 (Web Text))

Max Activating Token Index: 851

Text #8

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2649. Min Act: -0.1700

Data Index: 217762 (C4 (Web Text))

Max Activating Token Index: 522

Text #9

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.1712. Min Act: -0.1700

Data Index: 858959 (C4 (Web Text))

Max Activating Token Index: 505

Text #10

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2024. Min Act: -0.1700

Data Index: 1062413 (C4 (Web Text))

Max Activating Token Index: 67

Text #11

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2337. Min Act: -0.1700

Data Index: 960710 (C4 (Web Text))

Max Activating Token Index: 539

Text #12

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2337. Min Act: -0.1700

Data Index: 460764 (C4 (Web Text))

Max Activating Token Index: 999

Text #13

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2024. Min Act: -0.1700

Data Index: 342686 (C4 (Web Text))

Max Activating Token Index: 54

Text #14

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2024. Min Act: -0.1700

Data Index: 906077 (C4 (Web Text))

Max Activating Token Index: 837

Text #15

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.2337. Min Act: -0.1700

Data Index: 565238 (C4 (Web Text))

Max Activating Token Index: 999

Text #16

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.1712. Min Act: -0.1700

Data Index: 1329413 (C4 (Web Text))

Max Activating Token Index: 908

Text #17

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.1712. Min Act: -0.1700

Data Index: 1110332 (C4 (Web Text))

Max Activating Token Index: 738

Text #18

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.1399. Min Act: -0.1700

Data Index: 59589 (C4 (Web Text))

Max Activating Token Index: 887

Text #19

Max Range: 5.4524. Min Range: -5.4524

Max Act: 5.1087. Min Act: -0.1700

Data Index: 334261 (C4 (Web Text))

Max Activating Token Index: 85