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

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

Neuron 2015 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: 5.0506. Min Range: -5.0506

Max Act: 5.0506. Min Act: -0.1700

Data Index: 755048 (C4 (Web Text))

Max Activating Token Index: 113

Text #1

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.8006. Min Act: -0.1700

Data Index: 793794 (C4 (Web Text))

Max Activating Token Index: 65

Text #2

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.8318. Min Act: -0.1700

Data Index: 702201 (C4 (Web Text))

Max Activating Token Index: 583

Text #3

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.8006. Min Act: -0.1700

Data Index: 920776 (C4 (Web Text))

Max Activating Token Index: 303

Text #4

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.7693. Min Act: -0.1700

Data Index: 303923 (C4 (Web Text))

Max Activating Token Index: 939

Text #5

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.7068. Min Act: -0.1700

Data Index: 1154724 (C4 (Web Text))

Max Activating Token Index: 377

Text #6

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.7068. Min Act: -0.1700

Data Index: 344311 (C4 (Web Text))

Max Activating Token Index: 666

Text #7

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.7068. Min Act: -0.1700

Data Index: 1288891 (C4 (Web Text))

Max Activating Token Index: 876

Text #8

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.6756. Min Act: -0.1700

Data Index: 1340442 (C4 (Web Text))

Max Activating Token Index: 441

Text #9

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.6756. Min Act: -0.1700

Data Index: 838312 (C4 (Web Text))

Max Activating Token Index: 37

Text #10

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.6131. Min Act: -0.1700

Data Index: 869442 (C4 (Web Text))

Max Activating Token Index: 869

Text #11

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.5818. Min Act: -0.1700

Data Index: 781549 (C4 (Web Text))

Max Activating Token Index: 455

Text #12

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.6443. Min Act: -0.1700

Data Index: 546716 (C4 (Web Text))

Max Activating Token Index: 380

Text #13

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.6131. Min Act: -0.1700

Data Index: 557495 (C4 (Web Text))

Max Activating Token Index: 87

Text #14

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.5818. Min Act: -0.1700

Data Index: 641544 (C4 (Web Text))

Max Activating Token Index: 787

Text #15

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.6131. Min Act: -0.1700

Data Index: 181665 (C4 (Web Text))

Max Activating Token Index: 526

Text #16

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.5818. Min Act: -0.1700

Data Index: 970135 (C4 (Web Text))

Max Activating Token Index: 172

Text #17

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.6131. Min Act: -0.1700

Data Index: 16173 (C4 (Web Text))

Max Activating Token Index: 368

Text #18

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.6443. Min Act: -0.1700

Data Index: 7992 (C4 (Web Text))

Max Activating Token Index: 301

Text #19

Max Range: 5.0506. Min Range: -5.0506

Max Act: 4.6131. Min Act: -0.1700

Data Index: 367090 (C4 (Web Text))

Max Activating Token Index: 502