Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 103 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2920. Min Act: -0.1700

Data Index: 546048 (C4 (Web Text))

Max Activating Token Index: 888

Text #1

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2920. Min Act: -0.1700

Data Index: 1147956 (C4 (Web Text))

Max Activating Token Index: 565

Text #2

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2763. Min Act: -0.1700

Data Index: 758866 (C4 (Web Text))

Max Activating Token Index: 423

Text #3

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2605. Min Act: -0.1699

Data Index: 988945 (C4 (Web Text))

Max Activating Token Index: 594

Text #4

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2134. Min Act: -0.1700

Data Index: 93243 (C4 (Web Text))

Max Activating Token Index: 481

Text #5

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2291. Min Act: -0.1700

Data Index: 91471 (C4 (Web Text))

Max Activating Token Index: 328

Text #6

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2291. Min Act: -0.1700

Data Index: 403525 (C4 (Web Text))

Max Activating Token Index: 4

Text #7

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2134. Min Act: -0.1700

Data Index: 270679 (C4 (Web Text))

Max Activating Token Index: 621

Text #8

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2291. Min Act: -0.1700

Data Index: 546903 (C4 (Web Text))

Max Activating Token Index: 727

Text #9

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.1976. Min Act: -0.1700

Data Index: 762338 (C4 (Web Text))

Max Activating Token Index: 809

Text #10

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2134. Min Act: -0.1700

Data Index: 1609707 (Python Code)

Max Activating Token Index: 274

Text #11

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.1976. Min Act: -0.1700

Data Index: 757728 (C4 (Web Text))

Max Activating Token Index: 834

Text #12

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.2134. Min Act: -0.1700

Data Index: 673023 (C4 (Web Text))

Max Activating Token Index: 841

Text #13

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.1976. Min Act: -0.1700

Data Index: 1195501 (C4 (Web Text))

Max Activating Token Index: 700

Text #14

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.1662. Min Act: -0.1700

Data Index: 503548 (C4 (Web Text))

Max Activating Token Index: 953

Text #15

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.1819. Min Act: -0.1700

Data Index: 971042 (C4 (Web Text))

Max Activating Token Index: 506

Text #16

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.1662. Min Act: -0.1700

Data Index: 664422 (C4 (Web Text))

Max Activating Token Index: 813

Text #17

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.1819. Min Act: -0.1700

Data Index: 1611006 (Python Code)

Max Activating Token Index: 13

Text #18

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.1662. Min Act: -0.1700

Data Index: 272889 (C4 (Web Text))

Max Activating Token Index: 530

Text #19

Max Range: 3.2920. Min Range: -3.2920

Max Act: 3.1662. Min Act: -0.1700

Data Index: 464654 (C4 (Web Text))

Max Activating Token Index: 725