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

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

Neuron 564 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: 7.2788. Min Range: -7.2788

Max Act: 7.2788. Min Act: -0.1700

Data Index: 430723 (C4 (Web Text))

Max Activating Token Index: 148

Text #1

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.9350. Min Act: -0.1700

Data Index: 1228318 (C4 (Web Text))

Max Activating Token Index: 252

Text #2

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.8413. Min Act: -0.1700

Data Index: 758519 (C4 (Web Text))

Max Activating Token Index: 712

Text #3

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.8725. Min Act: -0.1700

Data Index: 824811 (C4 (Web Text))

Max Activating Token Index: 431

Text #4

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.8725. Min Act: -0.1700

Data Index: 639941 (C4 (Web Text))

Max Activating Token Index: 602

Text #5

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.8413. Min Act: -0.1700

Data Index: 849682 (C4 (Web Text))

Max Activating Token Index: 646

Text #6

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.8413. Min Act: -0.1700

Data Index: 412573 (C4 (Web Text))

Max Activating Token Index: 541

Text #7

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.9038. Min Act: -0.1700

Data Index: 1294398 (C4 (Web Text))

Max Activating Token Index: 342

Text #8

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.7788. Min Act: -0.1700

Data Index: 1449955 (Python Code)

Max Activating Token Index: 619

Text #9

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.7475. Min Act: -0.1700

Data Index: 576615 (C4 (Web Text))

Max Activating Token Index: 119

Text #10

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.7475. Min Act: -0.1700

Data Index: 1480222 (Python Code)

Max Activating Token Index: 552

Text #11

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.7788. Min Act: -0.1700

Data Index: 1145266 (C4 (Web Text))

Max Activating Token Index: 460

Text #12

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.7788. Min Act: -0.1700

Data Index: 643938 (C4 (Web Text))

Max Activating Token Index: 765

Text #13

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.7475. Min Act: -0.1700

Data Index: 492210 (C4 (Web Text))

Max Activating Token Index: 759

Text #14

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.6850. Min Act: -0.1700

Data Index: 136661 (C4 (Web Text))

Max Activating Token Index: 574

Text #15

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.6538. Min Act: -0.1700

Data Index: 100359 (C4 (Web Text))

Max Activating Token Index: 737

Text #16

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.6538. Min Act: -0.1700

Data Index: 221568 (C4 (Web Text))

Max Activating Token Index: 845

Text #17

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.7788. Min Act: -0.1700

Data Index: 928625 (C4 (Web Text))

Max Activating Token Index: 207

Text #18

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.6850. Min Act: -0.1700

Data Index: 1482030 (Python Code)

Max Activating Token Index: 569

Text #19

Max Range: 7.2788. Min Range: -7.2788

Max Act: 6.7475. Min Act: -0.1700

Data Index: 157821 (C4 (Web Text))

Max Activating Token Index: 566