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

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

Neuron 17 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.2969. Min Act: -0.1700

Data Index: 531056 (C4 (Web Text))

Max Activating Token Index: 1004

Text #1

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.1407. Min Act: -0.1700

Data Index: 704892 (C4 (Web Text))

Max Activating Token Index: 54

Text #2

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.2344. Min Act: -0.1700

Data Index: 434100 (C4 (Web Text))

Max Activating Token Index: 842

Text #3

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.1719. Min Act: -0.1700

Data Index: 890134 (C4 (Web Text))

Max Activating Token Index: 812

Text #4

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.1094. Min Act: -0.1700

Data Index: 735316 (C4 (Web Text))

Max Activating Token Index: 524

Text #5

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.1719. Min Act: -0.1700

Data Index: 1062220 (C4 (Web Text))

Max Activating Token Index: 682

Text #6

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.1094. Min Act: -0.1700

Data Index: 1106441 (C4 (Web Text))

Max Activating Token Index: 343

Text #7

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.0782. Min Act: -0.1700

Data Index: 631913 (C4 (Web Text))

Max Activating Token Index: 794

Text #8

Max Range: 5.2969. Min Range: -5.2969

Max Act: 4.9844. Min Act: -0.1700

Data Index: 136084 (C4 (Web Text))

Max Activating Token Index: 223

Text #9

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.0782. Min Act: -0.1700

Data Index: 206671 (C4 (Web Text))

Max Activating Token Index: 949

Text #10

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.1094. Min Act: -0.1700

Data Index: 629456 (C4 (Web Text))

Max Activating Token Index: 252

Text #11

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.0469. Min Act: -0.1700

Data Index: 1095818 (C4 (Web Text))

Max Activating Token Index: 819

Text #12

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.0469. Min Act: -0.1700

Data Index: 215163 (C4 (Web Text))

Max Activating Token Index: 871

Text #13

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.0782. Min Act: -0.1700

Data Index: 280296 (C4 (Web Text))

Max Activating Token Index: 990

Text #14

Max Range: 5.2969. Min Range: -5.2969

Max Act: 5.0782. Min Act: -0.1700

Data Index: 1053467 (C4 (Web Text))

Max Activating Token Index: 639

Text #15

Max Range: 5.2969. Min Range: -5.2969

Max Act: 4.9844. Min Act: -0.1700

Data Index: 786465 (C4 (Web Text))

Max Activating Token Index: 939

Text #16

Max Range: 5.2969. Min Range: -5.2969

Max Act: 4.9844. Min Act: -0.1700

Data Index: 69981 (C4 (Web Text))

Max Activating Token Index: 999

Text #17

Max Range: 5.2969. Min Range: -5.2969

Max Act: 4.9844. Min Act: -0.1700

Data Index: 344373 (C4 (Web Text))

Max Activating Token Index: 780

Text #18

Max Range: 5.2969. Min Range: -5.2969

Max Act: 4.9844. Min Act: -0.1700

Data Index: 1122478 (C4 (Web Text))

Max Activating Token Index: 897

Text #19

Max Range: 5.2969. Min Range: -5.2969

Max Act: 4.9532. Min Act: -0.1699

Data Index: 320165 (C4 (Web Text))

Max Activating Token Index: 886