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

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

Neuron 805 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: 5.7781. Min Range: -5.7781

Max Act: 5.7781. Min Act: -0.1700

Data Index: 1344892 (C4 (Web Text))

Max Activating Token Index: 7

Text #1

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.7468. Min Act: -0.1700

Data Index: 692506 (C4 (Web Text))

Max Activating Token Index: 378

Text #2

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.6531. Min Act: -0.1700

Data Index: 910597 (C4 (Web Text))

Max Activating Token Index: 174

Text #3

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.6218. Min Act: -0.1700

Data Index: 1002470 (C4 (Web Text))

Max Activating Token Index: 9

Text #4

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.5437. Min Act: -0.1700

Data Index: 831179 (C4 (Web Text))

Max Activating Token Index: 893

Text #5

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.6062. Min Act: -0.1700

Data Index: 550637 (C4 (Web Text))

Max Activating Token Index: 475

Text #6

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.5593. Min Act: -0.1700

Data Index: 1330788 (C4 (Web Text))

Max Activating Token Index: 768

Text #7

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.5593. Min Act: -0.1700

Data Index: 907406 (C4 (Web Text))

Max Activating Token Index: 1008

Text #8

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.5749. Min Act: -0.1700

Data Index: 610944 (C4 (Web Text))

Max Activating Token Index: 49

Text #9

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.5437. Min Act: -0.1700

Data Index: 394080 (C4 (Web Text))

Max Activating Token Index: 705

Text #10

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.5281. Min Act: -0.1700

Data Index: 958615 (C4 (Web Text))

Max Activating Token Index: 885

Text #11

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.4656. Min Act: -0.1700

Data Index: 499215 (C4 (Web Text))

Max Activating Token Index: 876

Text #12

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.4499. Min Act: -0.1700

Data Index: 209052 (C4 (Web Text))

Max Activating Token Index: 354

Text #13

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.4968. Min Act: -0.1700

Data Index: 1350361 (C4 (Web Text))

Max Activating Token Index: 150

Text #14

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.4968. Min Act: -0.1700

Data Index: 104053 (C4 (Web Text))

Max Activating Token Index: 985

Text #15

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.4499. Min Act: -0.1700

Data Index: 19732 (C4 (Web Text))

Max Activating Token Index: 10

Text #16

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.4656. Min Act: -0.1700

Data Index: 924252 (C4 (Web Text))

Max Activating Token Index: 1019

Text #17

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.4343. Min Act: -0.1700

Data Index: 331254 (C4 (Web Text))

Max Activating Token Index: 908

Text #18

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.5124. Min Act: -0.1700

Data Index: 41585 (C4 (Web Text))

Max Activating Token Index: 834

Text #19

Max Range: 5.7781. Min Range: -5.7781

Max Act: 5.4812. Min Act: -0.1700

Data Index: 714634 (C4 (Web Text))

Max Activating Token Index: 7