Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1022 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.4702. Min Act: -0.1700

Data Index: 465447 (C4 (Web Text))

Max Activating Token Index: 1013

Text #1

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.3764. Min Act: -0.1700

Data Index: 103069 (C4 (Web Text))

Max Activating Token Index: 704

Text #2

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.2983. Min Act: -0.1700

Data Index: 1353687 (C4 (Web Text))

Max Activating Token Index: 806

Text #3

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.1889. Min Act: -0.1700

Data Index: 1173015 (C4 (Web Text))

Max Activating Token Index: 963

Text #4

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.2045. Min Act: -0.1700

Data Index: 820034 (C4 (Web Text))

Max Activating Token Index: 997

Text #5

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.1889. Min Act: -0.1700

Data Index: 543001 (C4 (Web Text))

Max Activating Token Index: 743

Text #6

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.0952. Min Act: -0.1700

Data Index: 1002971 (C4 (Web Text))

Max Activating Token Index: 654

Text #7

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.1733. Min Act: -0.1700

Data Index: 534376 (C4 (Web Text))

Max Activating Token Index: 685

Text #8

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.0327. Min Act: -0.1700

Data Index: 771387 (C4 (Web Text))

Max Activating Token Index: 754

Text #9

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.0170. Min Act: -0.1700

Data Index: 138586 (C4 (Web Text))

Max Activating Token Index: 659

Text #10

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.0795. Min Act: -0.1700

Data Index: 1473083 (Python Code)

Max Activating Token Index: 791

Text #11

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.0327. Min Act: -0.1700

Data Index: 1513074 (Python Code)

Max Activating Token Index: 766

Text #12

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.0014. Min Act: -0.1700

Data Index: 1523873 (Python Code)

Max Activating Token Index: 824

Text #13

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.0483. Min Act: -0.1700

Data Index: 330466 (C4 (Web Text))

Max Activating Token Index: 662

Text #14

Max Range: 8.4702. Min Range: -8.4702

Max Act: 8.0327. Min Act: -0.1700

Data Index: 462240 (C4 (Web Text))

Max Activating Token Index: 818

Text #15

Max Range: 8.4702. Min Range: -8.4702

Max Act: 7.9702. Min Act: -0.1700

Data Index: 1229526 (C4 (Web Text))

Max Activating Token Index: 711

Text #16

Max Range: 8.4702. Min Range: -8.4702

Max Act: 7.8920. Min Act: -0.1700

Data Index: 270593 (C4 (Web Text))

Max Activating Token Index: 989

Text #17

Max Range: 8.4702. Min Range: -8.4702

Max Act: 7.9858. Min Act: -0.1700

Data Index: 1457437 (Python Code)

Max Activating Token Index: 943

Text #18

Max Range: 8.4702. Min Range: -8.4702

Max Act: 7.9858. Min Act: -0.1700

Data Index: 1372926 (Python Code)

Max Activating Token Index: 736

Text #19

Max Range: 8.4702. Min Range: -8.4702

Max Act: 7.9545. Min Act: -0.1700

Data Index: 848036 (C4 (Web Text))

Max Activating Token Index: 604