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

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

Neuron 369 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: 5.1670. Min Range: -5.1670

Max Act: 5.1670. Min Act: -0.1700

Data Index: 21598 (C4 (Web Text))

Max Activating Token Index: 793

Text #1

Max Range: 5.1670. Min Range: -5.1670

Max Act: 5.1045. Min Act: -0.1700

Data Index: 217626 (C4 (Web Text))

Max Activating Token Index: 570

Text #2

Max Range: 5.1670. Min Range: -5.1670

Max Act: 5.1045. Min Act: -0.1700

Data Index: 506968 (C4 (Web Text))

Max Activating Token Index: 549

Text #3

Max Range: 5.1670. Min Range: -5.1670

Max Act: 5.0732. Min Act: -0.1700

Data Index: 441488 (C4 (Web Text))

Max Activating Token Index: 381

Text #4

Max Range: 5.1670. Min Range: -5.1670

Max Act: 5.1357. Min Act: -0.1700

Data Index: 885261 (C4 (Web Text))

Max Activating Token Index: 348

Text #5

Max Range: 5.1670. Min Range: -5.1670

Max Act: 5.0107. Min Act: -0.1700

Data Index: 1138028 (C4 (Web Text))

Max Activating Token Index: 986

Text #6

Max Range: 5.1670. Min Range: -5.1670

Max Act: 5.0732. Min Act: -0.1700

Data Index: 1076985 (C4 (Web Text))

Max Activating Token Index: 749

Text #7

Max Range: 5.1670. Min Range: -5.1670

Max Act: 5.0420. Min Act: -0.1700

Data Index: 750887 (C4 (Web Text))

Max Activating Token Index: 79

Text #8

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9795. Min Act: -0.1700

Data Index: 696644 (C4 (Web Text))

Max Activating Token Index: 555

Text #9

Max Range: 5.1670. Min Range: -5.1670

Max Act: 5.0420. Min Act: -0.1700

Data Index: 1036197 (C4 (Web Text))

Max Activating Token Index: 445

Text #10

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9795. Min Act: -0.1700

Data Index: 360608 (C4 (Web Text))

Max Activating Token Index: 569

Text #11

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9795. Min Act: -0.1700

Data Index: 858257 (C4 (Web Text))

Max Activating Token Index: 740

Text #12

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9482. Min Act: -0.1700

Data Index: 591738 (C4 (Web Text))

Max Activating Token Index: 159

Text #13

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9795. Min Act: -0.1700

Data Index: 1077185 (C4 (Web Text))

Max Activating Token Index: 593

Text #14

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9482. Min Act: -0.1700

Data Index: 1086178 (C4 (Web Text))

Max Activating Token Index: 918

Text #15

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9795. Min Act: -0.1700

Data Index: 910857 (C4 (Web Text))

Max Activating Token Index: 141

Text #16

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9482. Min Act: -0.1700

Data Index: 389840 (C4 (Web Text))

Max Activating Token Index: 142

Text #17

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9482. Min Act: -0.1700

Data Index: 1181812 (C4 (Web Text))

Max Activating Token Index: 740

Text #18

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9795. Min Act: -0.1700

Data Index: 406038 (C4 (Web Text))

Max Activating Token Index: 97

Text #19

Max Range: 5.1670. Min Range: -5.1670

Max Act: 4.9795. Min Act: -0.1700

Data Index: 541450 (C4 (Web Text))

Max Activating Token Index: 355