Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 695 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 4.6915. Min Range: -4.6915

Max Act: 4.6915. Min Act: -0.1686

Data Index: 224891 (C4 (Web Text))

Max Activating Token Index: 676

Text #1

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.7949. Min Act: -0.1696

Data Index: 711980 (C4 (Web Text))

Max Activating Token Index: 785

Text #2

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.8271. Min Act: -0.1699

Data Index: 1239722 (C4 (Web Text))

Max Activating Token Index: 732

Text #3

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.7440. Min Act: -0.1700

Data Index: 63495 (C4 (Web Text))

Max Activating Token Index: 860

Text #4

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.7225. Min Act: -0.1700

Data Index: 262210 (C4 (Web Text))

Max Activating Token Index: 98

Text #5

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.6787. Min Act: -0.1686

Data Index: 1547142 (Python Code)

Max Activating Token Index: 824

Text #6

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.6843. Min Act: -0.1686

Data Index: 1397968 (Python Code)

Max Activating Token Index: 445

Text #7

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.6542. Min Act: -0.1686

Data Index: 469704 (C4 (Web Text))

Max Activating Token Index: 875

Text #8

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.6221. Min Act: -0.1699

Data Index: 661997 (C4 (Web Text))

Max Activating Token Index: 629

Text #9

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.6312. Min Act: -0.1686

Data Index: 1556416 (Python Code)

Max Activating Token Index: 728

Text #10

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5962. Min Act: -0.1686

Data Index: 1471448 (Python Code)

Max Activating Token Index: 32

Text #11

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5993. Min Act: -0.1697

Data Index: 146690 (C4 (Web Text))

Max Activating Token Index: 1011

Text #12

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5687. Min Act: -0.1686

Data Index: 307505 (C4 (Web Text))

Max Activating Token Index: 92

Text #13

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5806. Min Act: -0.1699

Data Index: 1444782 (Python Code)

Max Activating Token Index: 225

Text #14

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5745. Min Act: -0.1700

Data Index: 1617576 (Python Code)

Max Activating Token Index: 245

Text #15

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5970. Min Act: -0.1700

Data Index: 1445301 (Python Code)

Max Activating Token Index: 619

Text #16

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5961. Min Act: -0.1697

Data Index: 450907 (C4 (Web Text))

Max Activating Token Index: 962

Text #17

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5488. Min Act: -0.1700

Data Index: 702067 (C4 (Web Text))

Max Activating Token Index: 169

Text #18

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5462. Min Act: -0.1686

Data Index: 369208 (C4 (Web Text))

Max Activating Token Index: 949

Text #19

Max Range: 4.6915. Min Range: -4.6915

Max Act: 3.5417. Min Act: -0.1698

Data Index: 307859 (C4 (Web Text))

Max Activating Token Index: 646