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

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

Neuron 2015 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.9415. Min Act: -0.1700

Data Index: 114782 (C4 (Web Text))

Max Activating Token Index: 750

Text #1

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.8946. Min Act: -0.1700

Data Index: 183718 (C4 (Web Text))

Max Activating Token Index: 267

Text #2

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.8477. Min Act: -0.1700

Data Index: 1192745 (C4 (Web Text))

Max Activating Token Index: 560

Text #3

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7695. Min Act: -0.1700

Data Index: 670679 (C4 (Web Text))

Max Activating Token Index: 599

Text #4

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.8008. Min Act: -0.1700

Data Index: 574842 (C4 (Web Text))

Max Activating Token Index: 877

Text #5

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7695. Min Act: -0.1700

Data Index: 449120 (C4 (Web Text))

Max Activating Token Index: 235

Text #6

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.8320. Min Act: -0.1700

Data Index: 669921 (C4 (Web Text))

Max Activating Token Index: 1017

Text #7

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7851. Min Act: -0.1700

Data Index: 737210 (C4 (Web Text))

Max Activating Token Index: 380

Text #8

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7851. Min Act: -0.1700

Data Index: 1256644 (C4 (Web Text))

Max Activating Token Index: 504

Text #9

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7538. Min Act: -0.1700

Data Index: 327840 (C4 (Web Text))

Max Activating Token Index: 494

Text #10

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7695. Min Act: -0.1700

Data Index: 775961 (C4 (Web Text))

Max Activating Token Index: 144

Text #11

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7069. Min Act: -0.1700

Data Index: 926418 (C4 (Web Text))

Max Activating Token Index: 828

Text #12

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7225. Min Act: -0.1700

Data Index: 1100568 (C4 (Web Text))

Max Activating Token Index: 883

Text #13

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.6913. Min Act: -0.1700

Data Index: 1010925 (C4 (Web Text))

Max Activating Token Index: 558

Text #14

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7069. Min Act: -0.1700

Data Index: 837268 (C4 (Web Text))

Max Activating Token Index: 582

Text #15

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.6756. Min Act: -0.1700

Data Index: 725692 (C4 (Web Text))

Max Activating Token Index: 631

Text #16

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.7069. Min Act: -0.1700

Data Index: 314641 (C4 (Web Text))

Max Activating Token Index: 384

Text #17

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.6913. Min Act: -0.1700

Data Index: 1263998 (C4 (Web Text))

Max Activating Token Index: 123

Text #18

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.6286. Min Act: -0.1700

Data Index: 662666 (C4 (Web Text))

Max Activating Token Index: 780

Text #19

Max Range: 3.9415. Min Range: -3.9415

Max Act: 3.6443. Min Act: -0.1700

Data Index: 741471 (C4 (Web Text))

Max Activating Token Index: 249