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

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

Neuron 346 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.7970. Min Act: -0.1700

Data Index: 676032 (C4 (Web Text))

Max Activating Token Index: 198

Text #1

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.7970. Min Act: -0.1700

Data Index: 579835 (C4 (Web Text))

Max Activating Token Index: 373

Text #2

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.7657. Min Act: -0.1700

Data Index: 721868 (C4 (Web Text))

Max Activating Token Index: 92

Text #3

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.6248. Min Act: -0.1700

Data Index: 1655200 (Python Code)

Max Activating Token Index: 562

Text #4

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.5622. Min Act: -0.1700

Data Index: 19989 (C4 (Web Text))

Max Activating Token Index: 602

Text #5

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.5935. Min Act: -0.1700

Data Index: 88308 (C4 (Web Text))

Max Activating Token Index: 301

Text #6

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.5465. Min Act: -0.1700

Data Index: 1346573 (C4 (Web Text))

Max Activating Token Index: 76

Text #7

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.5622. Min Act: -0.1700

Data Index: 1332692 (C4 (Web Text))

Max Activating Token Index: 250

Text #8

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.5465. Min Act: -0.1700

Data Index: 902549 (C4 (Web Text))

Max Activating Token Index: 323

Text #9

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.5465. Min Act: -0.1700

Data Index: 259747 (C4 (Web Text))

Max Activating Token Index: 573

Text #10

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.5152. Min Act: -0.1700

Data Index: 904275 (C4 (Web Text))

Max Activating Token Index: 338

Text #11

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.4525. Min Act: -0.1700

Data Index: 1221186 (C4 (Web Text))

Max Activating Token Index: 111

Text #12

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.4682. Min Act: -0.1700

Data Index: 588170 (C4 (Web Text))

Max Activating Token Index: 590

Text #13

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.4212. Min Act: -0.1700

Data Index: 1275593 (C4 (Web Text))

Max Activating Token Index: 288

Text #14

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.4525. Min Act: -0.1700

Data Index: 516864 (C4 (Web Text))

Max Activating Token Index: 57

Text #15

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.4212. Min Act: -0.1700

Data Index: 752844 (C4 (Web Text))

Max Activating Token Index: 45

Text #16

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.4525. Min Act: -0.1700

Data Index: 636123 (C4 (Web Text))

Max Activating Token Index: 605

Text #17

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.4369. Min Act: -0.1700

Data Index: 1059558 (C4 (Web Text))

Max Activating Token Index: 633

Text #18

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.3741. Min Act: -0.1700

Data Index: 1590223 (Python Code)

Max Activating Token Index: 132

Text #19

Max Range: 3.7970. Min Range: -3.7970

Max Act: 3.4055. Min Act: -0.1700

Data Index: 1486054 (Python Code)

Max Activating Token Index: 243