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

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

Neuron 454 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: 5.2961. Min Range: -5.2961

Max Act: 5.2649. Min Act: -0.1700

Data Index: 941324 (C4 (Web Text))

Max Activating Token Index: 968

Text #1

Max Range: 5.2961. Min Range: -5.2961

Max Act: 5.2961. Min Act: -0.1700

Data Index: 925298 (C4 (Web Text))

Max Activating Token Index: 116

Text #2

Max Range: 5.2961. Min Range: -5.2961

Max Act: 5.0149. Min Act: -0.1700

Data Index: 154195 (C4 (Web Text))

Max Activating Token Index: 825

Text #3

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.9211. Min Act: -0.1700

Data Index: 45479 (C4 (Web Text))

Max Activating Token Index: 932

Text #4

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.8899. Min Act: -0.1700

Data Index: 51948 (C4 (Web Text))

Max Activating Token Index: 992

Text #5

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.9524. Min Act: -0.1700

Data Index: 897086 (C4 (Web Text))

Max Activating Token Index: 258

Text #6

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.8586. Min Act: -0.1700

Data Index: 478065 (C4 (Web Text))

Max Activating Token Index: 860

Text #7

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.8899. Min Act: -0.1700

Data Index: 315022 (C4 (Web Text))

Max Activating Token Index: 923

Text #8

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.7961. Min Act: -0.1700

Data Index: 222006 (C4 (Web Text))

Max Activating Token Index: 652

Text #9

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.8274. Min Act: -0.1700

Data Index: 591810 (C4 (Web Text))

Max Activating Token Index: 269

Text #10

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.7961. Min Act: -0.1700

Data Index: 1186850 (C4 (Web Text))

Max Activating Token Index: 391

Text #11

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.7649. Min Act: -0.1700

Data Index: 735522 (C4 (Web Text))

Max Activating Token Index: 1013

Text #12

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.7649. Min Act: -0.1700

Data Index: 178207 (C4 (Web Text))

Max Activating Token Index: 1003

Text #13

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.8586. Min Act: -0.1700

Data Index: 931915 (C4 (Web Text))

Max Activating Token Index: 644

Text #14

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.8274. Min Act: -0.1700

Data Index: 940971 (C4 (Web Text))

Max Activating Token Index: 356

Text #15

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.7961. Min Act: -0.1700

Data Index: 500095 (C4 (Web Text))

Max Activating Token Index: 447

Text #16

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.7649. Min Act: -0.1700

Data Index: 242153 (C4 (Web Text))

Max Activating Token Index: 394

Text #17

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.7024. Min Act: -0.1700

Data Index: 145123 (C4 (Web Text))

Max Activating Token Index: 246

Text #18

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.7649. Min Act: -0.1700

Data Index: 946506 (C4 (Web Text))

Max Activating Token Index: 80

Text #19

Max Range: 5.2961. Min Range: -5.2961

Max Act: 4.7649. Min Act: -0.1700

Data Index: 222541 (C4 (Web Text))

Max Activating Token Index: 658